Potassium Changing from Pro- to Anti-Convulsant in the Epileptic Juvenile Rat Hippocampus

by

Wilson Jonathan Yu

A thesis submitted in conformity with the requirements for the degree of Master of Science

Graduate Department of Physiology University of Toronto

© Copyright by Wilson Jonathan Yu 2009

ii

Potassium Changing from Pro- to Anti-Convulsant in the Epileptic Juvenile Rat Hippocampus

Degree of Master of Science, 2009

Wilson Jonathan Yu

Graduate Department of Physiology

University of Toronto

Abstract

Elevations in extracellular potassium (K+

e) accompany seizure-like events (SLEs), but

elevated K+ may also participate in seizure cessation. The objective of this thesis was to

investigate the possibility that K+ may undergo a pro- to anti-convulsant switch in the epileptic

juvenile (postnatal day 17-21) rat hippocampus.

Field recordings were performed in the CA1 pyramidal layer. SLEs and primary

afterdischarges (PADs) were induced with 0.25 mM Mg/5 mM K+

perfusion or tetanic

stimulation of the Schaffer collaterals respectively. In these seizure models, elevating [K+]e

beyond 7.5 mM showed anticonvulsant properties. The addition of ZD7288, a blocker of the

hyperpolarization activated nonspecific cationic current (Ih) and allowed SLEs to continue even

in elevated [K+]e. This suggests that [K

+]e switches from being pro- to anti-convulsant, in part

due to an elevated [K+]e-induced potentiation of Ih. Ih likely contributes to this anticonvulsant

behavior by decreasing membrane resistance and subsequently attenuating summation of

incoming EPSPs.

iii

Acknowledgements

Many of the people that I met and worked with for the duration of my Masters of Science

(MSc) degree have contributed greatly to making the past few years a very rewarding learning

experience.

First, I would like to thank my professor, Dr. Peter Carlen, for his supervision and for

granting me the privilege of working in his lab. His unquenchable curiosity and tireless drive to

uncover more in the scientific field are truly inspirational.

I would also like to thank Dr. Damian Shin for his gracious and continued mentorship,

without which I would not be the person I am today. His patience in teaching and helping all

those around him, both in and out of lab, truly sets an example for me and my conduct in the

future.

Additionally, I would like to extend thanks to my supervisory committee, Drs. Liang Zhang,

Mac Burnham, and Evelyn Lambe, for their advice and feedback throughout my MSc. Under

their guidance, I learned much more regarding the essence required to be a successful scientist.

Furthermore, I would like to thank many of my labmates, in particular Demitre Serletis, Joe

El-Hayek, Marina Samoilova, as well as my other friends, and my family for their support and

encouragement throughout my academic journey.

Last, but not least, I would like to thank the Canadian Institute of Health Research (CIHR)

for their funding of this project, as well as the support from the Toronto Western Research

Institute (TWRI) at the Toronto Western Hospital (TWH), the University of Toronto Department

of Physiology, the Collaborative Program in Neuroscience, and the University of Toronto

Epilepsy Research Program (UTERP).

iv

Table of Contents Abstract ........................................................................................................................................... ii

Acknowledgements ........................................................................................................................ iii

Table of Contents ........................................................................................................................... iv

List of Tables .................................................................................................................................. v

List of Figures ................................................................................................................................ vi

List of Abbreviations .................................................................................................................... vii

1 A Brief Review of Epilepsy..................................................................................................... 1

1.1 Introduction ...................................................................................................................... 1

1.2 Clinical Types of Epilepsy ............................................................................................... 2

1.3 Clinical Electrophysiology of Seizures ............................................................................ 3

1.4 Cellular Mechanisms of Epilepsy .................................................................................... 5

1.5 Anticonvulsant Treatment Options ................................................................................ 10

2 Potassium and Epilepsy ......................................................................................................... 14

2.1 Introduction .................................................................................................................... 14

2.2 Basic Membrane Physiology .......................................................................................... 14

2.3 The Role of Potassium in Seizure Activity .................................................................... 16

3 The Ih Current ....................................................................................................................... 21

3.1 Introduction .................................................................................................................... 21

3.2 HCN Channel Structure ................................................................................................. 21

3.3 HCN Channel Expression .............................................................................................. 23

3.4 Functional Properties of Ih ............................................................................................. 25

3.5 Ih and Epilepsy ............................................................................................................... 26

4 Research Rationale, Objectives, and Hypotheses .................................................................. 28

5 Materials and Methods .......................................................................................................... 29

6 Results ................................................................................................................................... 33

6.1 The Effect of Elevating [K+]e to Ceiling Levels Observed During SLEs ..................... 33

6.2 Finding the Anticonvulsant Range of [K+]e in the Low Mg

2+/5 mM K

+ Model ............ 36

6.3 The Effects of Elevating [K+]e in the Tetanization Model ............................................. 40

6.4 The Effect of Blocking Ih on K+-induced attenuation of SLEs ..................................... 42

7 Discussion .............................................................................................................................. 44

8 Conclusions ........................................................................................................................... 50

9 Ongoing/Future Work............................................................................................................ 51

10 References .......................................................................................................................... 53

v

List of Tables

Table 1. Agents that produce focal epilepsy and their mechanisms of action. .............................. 7

vi

List of Figures

Figure 1. The Potassium Hypothesis of Seizure Generation. ...................................................... 17 Figure 2. Proposed Effects by which Elevated K

+ Triggers SLEs. ............................................. 19

Figure 3. Evoked responses and spontaneous potentials in aCSF and low Mg2+

/5 mM K+. ....... 34

Figure 4. Evoked responses and spontaneous potentials in low Mg2+

/12.5 mM K+ and low

Mg2+

/10 mM K+. ........................................................................................................................... 35

Figure 5. Evoked responses and spontaneous potentials in low Mg2+

/7.5 mM K+, low Mg

2+/6.5

mM K+, and low Mg

2+/6 mM K

+. ................................................................................................. 38

Figure 6. Normalized group data showing the effects of elevating [K+]e on the population spike

and fEPSP amplitude, and the SLE amplitude and duration in the low Mg2+

/5 mM K+ model. .. 39

Figure 7. Evoked responses and tetanization-induced PADs in 2.5 mM K+, 5 mM K

+, and 7.5

mM K+. ......................................................................................................................................... 41

Figure 8. The Effect of Blocking Ih on K+-induced Attenuation of SLEs. ................................. 43

vii

List of Abbreviations

aCSF artificial cerebral spinal fluid

AEDs antiepileptic drugs

ATP adenosine triphosphate

AMPA α-amino-3-hydroxy-5-methylisoxazole-4- propionic acid

BBB blood-brain barrier

CA1 cornu ammonis area 1

CA3 cornu ammonis area 3

cAMP cyclic adenosine monophosphate

cGMP cyclic guanosine monophosphate

CNBD cyclic nucleotide binding domain

CNG cyclic nucleotide gated

DBS deep brain stimulation

DS depolarization shift

EEG electroencephalogram

fEPSP field excitatory postsynaptic potential

GABA gamma aminobutyric acid

GAERs genetic absence epilepsy rats

h∞ steady state inactivation

HCN hyperpolarization-activated and cyclic nucleotide gated

If funny current

Ih hyperpolarization-activated nonspecific cationic current

IK,A transient A potassium current

viii

IK,DR delayed rectifier potassium current

INa,P persistent sodium current

INa,T transient sodium current

IPSP inhibitory postsynaptic potential

Iq queer current

[K+]e extracellular K

+

LTD long term depression

LTP long term potentiation

LVA low voltage activated

NMDARs N-methyl-D-aspartate receptors

NRSF neuron restrictive silencing factor

O/A stratum oriens and alveus

PADs primary afterdischarges

PTZ pentylenetetrazol

Rm membrane resistance

RMP resting membrane potential

SLEs seizure-like events

SW spike and wave

VDCCs voltage-dependent calcium channels

VDSCs voltage-dependent sodium channels

V1/2 half maximal activation potential

Vm membrane potential

ZD7288 4-ethylphenylamino-1,2-dimethyl-6-methylaminopyrimidinium chloride

1

1 A Brief Review of Epilepsy

1.1 Introduction

The brain is the most complex organ in the body and in order for it to function appropriately,

the many billions of individual neuronal and glial cells that make up this system must work

together in a coordinated fashion. Pathologies arise when this balance is disturbed, and one such

condition is epilepsy. The epilepsies are a group of disorders characterized by spontaneous,

recurrent seizure activity, seizures being defined as sustained periods of hyperexcitation and/or

hypersynchrony associated with brain dysfunction (Hauser et al. 1993). There are approximately

1% of people who have epilepsy at any given time (Hauser et al. 1993). Epilepsy is considered

one of the most serious neurological disorders, as it heavily affects the quality of life of afflicted

individuals, and many epileptic patients have intractable seizures, so their condition cannot

successfully be controlled by current drugs. As such, further research must be conducted to

improve our understanding of the epileptic condition and to examine other treatment options.

This chapter will attempt to briefly describe what is currently known regarding epilepsy, and the

possible mechanisms and treatments thereof.

2

1.2 Clinical Types of Epilepsy

Epilepsy itself, as a group of disorders, can be most broadly categorized as symptomatic

or essential (Lüders and Noachtar 2000). Symptomatic epilepsy refers to seizure disorders that

are a result of an identifiable problem, such as a brain lesion. Essential, or idiopathic epilepsy, is

the opposite, where the seizure disorders have an unknown origin, though many idiopathic

epilepsies are now being found to have a genetic basis (Singh et al. 2002).

Seizures can further be classified as convulsive or nonconvulsive, and even more

specifically, as generalized or partial, depending on the area(s) of the brain affected (Somjen

2004). Convulsive seizures are generalized tonic-clonic or “grand mal” seizures. These seizures

have two phases: the initial tonic phase, involving simultaneous spasms, followed by the clonic

phase, where the body undergoes rhythmic jerking movements (Engel 1989). They are

considered generalized, since the seizures affect both cerebral hemispheres significantly and

there is a loss of consciousness. If the seizure persists for an extended period of time or occurs

so frequently that there is no recovery period in between, then it is referred to as a status

epilepticus. Another form of generalized seizure is referred to as an absence or “petit mal”

seizure, where there is a momentary lapse of consciousness, but no major convulsions (Cortez et

al. 2001).

There can also be partial seizures, if only a limited area of the brain is affected (Engel

1989). These seizures, while limited to that area, are known as focal seizures, but seizures can

propagate and evolve into secondary generalized seizures. Partial seizures can further be broken

down into simple or complex partial seizures, depending on whether consciousness is retained, in

the former case, or if consciousness is lost of impaired, as in the latter case. Simple partial

seizures can arise from the neocortical area, and when they affect the motor cortex, they can

3

cause convulsions, known as Jacksonian fits, and sometimes subsequent temporary paralysis in

the muscle groups associated with the affected motor area (Lüders and Noachtar 2000). On the

other hand, complex partial seizures affect the conscious state and tend to originate from

subcortical areas in the temporal lobe, in particular, the amygdala and hippocampus. As such,

complex partial seizures are grouped under the condition known as temporal lobe epilepsy.

These seizures are characterized by automatisms, which are stereotyped behaviours resulting

from the mobilization of specific neuronal sequences (Somjen 2004).

In summary, there are many different classifications for the array of seizure disorders

known as epilepsy, and this introduction only briefly touches upon some of the main forms of

epilepsy.

1.3 Clinical Electrophysiology of Seizures

As seizures are electrical disturbances which disrupt the normal activity of the brain,

seizure-like events (SLEs) can be readily recorded through the use of electrophysiological

techniques. On the clinical level, seizures are studied mostly with electroencephalography

(EEG) (Niedermeyer 1972). The EEG is a product of synaptic activity and represents rhythmic

oscillations in voltage due to the synchronized firing of postsynaptic potentials from neurons

around the electrode positions. The activity recorded in EEG recordings can be split into a

number of frequency ranges: delta (0.1 – 3.5 Hz), theta (4 – 7.5 Hz), alpha (8-13 Hz), beta (14-30

Hz), and gamma (30-100 Hz), and seizures are typically dominated by high frequency activity

(Niedermeyer 1972).

Seizure (ictal) events can be easily observed on EEG recordings, as they are the result of

hypersynchronization of many cells. For generalized tonic seizures, the EEG recordings reveals

4

high-voltage, fast-discharge pattern, with a frequency typically around 15-20 Hz, though this

frequency range is not always the case (Lüders and Noachtar 2000). Tonic phases are often

followed by clonus, where the EEG shows large amplitude, high frequency bursting activity

corresponding to the jerks seen in affected muscles. After the ictal episode, there is a

characteristic period of postictal silence (Gastaut and Broughton 1972).

Absence seizures also show a standard pattern of activity, whereby there are complex

waves that occur at a rhythm of about 3 Hz, and between the waves, there is a spike component

(Walton 1985). As such, the standard discharge characterizing absence seizures is a 3 per second

spike-and-wave (SW) pattern. The waves are due to fluctuations in membrane potentials, and

the spikes are a result of the depolarizing phase of action potentials.

Apart from ictal activity, EEGs can also record interictal activity, which is the activity

between seizures. These show up in the form of polyspike complexes, large-amplitude slow

waves, isolated spike-wave complexes, and sharp waves. Spikes and sharp waves are

differentiated on the EEG by timing, since sharp waves can last up to 200 ms, whereas spikes

only last about 70 ms (Engel 1989). The role or function of interictal activity is still debated.

Some have suggested that these interictal events may be suggestive of an upcoming ictal event,

as some studies do show a progression of interictal activity leading to full ictal discharges (Jami

1972; Prince et al. 1983; Jensen and Yaari 1988; Amzica and Steriade 2000). Others propose

that interictal activity may reflect past seizures as opposed to indicating future seizures (Gotman

1991). More recently, it has been suggested that interictal and ictal activity may actually be

generated by different cell groups, and interictal activity may be inhibitory and actually

protective against seizures (Swartzwelder et al. 1987; de Curtis et al. 1999; Avoli 2001).

5

Somjen suggests that similarities in the interictal activity electrically may not necessarily mean

that all interictal activity is driven by the same mechanisms and processes, and this may explain

why there is conflicting evidence for the role of interictal activity in the promotion or protection

of ictal events (Somjen 2004).

1.4 Cellular Mechanisms of Epilepsy

Over the past number of decades, there has been a steady progression in technology, and

this has allowed for research into the cellular and molecular mechanisms behind seizures. As

mentioned in section 1.3, on the EEG, seizures can be seen as ictal events, but the presence of

interictal discharges is also indicative of a hyperexcitable area that has potential for

epileptogenicity. The connection between interictal discharges and seizures seems to be most

closely related in focal seizure models (Dichter and Ayala 1987).

In the focal seizure models, during interictal discharges, there is a considerable

depolarization shift (DS) with a burst of action potentials superimposed on the wave of

depolarization, followed by a post-DS hyperpolarization where neurons are inhibited (Dichter

and Spencer 1969; Lopantsev and Avoli 1998). As the seizure-like activity develops, the post-

DS hyperpolarization shortens, and eventually it is replaced by depolarizing waves (Ayala et al.

1970). As the interictal discharges continue to build, there is a steady increase in extracellular

K+ as well as a decrease of Ca

2+ as a result of intense neuronal activity (Fertziger and Ranck

1970), but in areas surrounding the focus, there is general inhibition. On the EEG, after

successive interictal discharges, afterdischarges develop and eventually grow into a full blown

seizure event.

6

At this point, nearby areas of the brain may be recruited into the seizure process and so the

seizure may spread and become generalized (Dichter and Ayala 1987). After the seizure episode

stops, there is a post-ictal depression marked by hyperpolarization of the neuronal membrane.

To study the mechanisms behind epilepsy, naturally, experimental models are utilized.

What adds to the complexity of epilepsy is that various agents with unrelated mechanisms of

action can still provoke focal seizure-like events with the similar stereotyped DS

hyperpolarization sequence (Table 1) (Dichter and Ayala 1987). These agents seem to largely

have an effect on the intrinsic membrane properties and functions of cells. Thus, whether

seizures occur at any focal point depends on whether the normal function of that area is changed

due to alterations in the intrinsic currents of the cells in the region, or a shift in synaptic efficacy.

Factors that would then have to be taken into account for that epileptogenic region include the

density and location of channels in the area, possible interactions between different currents, the

general synaptic organization, as well as possible neuromodulators that may affect the currents

through second messenger pathways (Dichter and Ayala 1987)

Current studies support the idea that the DS is a result of excitatory synaptic currents,

which may be amplified by other intrinsic currents (Dichter and Spencer 1969; Ayala et al. 1970;

Johnston and Brown 1981). There are a number of proposed mechanisms for how the excitatory

synaptic currents can become enhanced enough to produce SLEs.

7

Convulsant Agent Proposed Cellular Mechanism Penicillin, bicuculline, picrotoxin Antagonist of GABAA receptors

Allylglycine, 3-mercaptoproprionic acid Block synthesis of GABA

Ammonium salts Block Cl- pump

Tetramethylammonium, 4-aminopyridine (4-

AP)

Block K+ currents

Barium Enhance Ca2+

currents and block K+ currents

Enkephalin Inhibit inhibitory interneurons

Elevated [K+]e Affect intrinsic membrane characteristics

Depolarize neurons

Increase emphatic coupling

Low [Ca2+

]e Increase neuronal excitability

Decrease neuronal threshold

Low [Mg2+

]e Reduce surface screening

Increase presynaptic transmitter release

Potentiate NMDAR activation

Cardiac glycosides Decrease Na+ pump

Increase [K+]e

Increase Ca2+

spikes

Alumina Cream Neuronal Loss

Selective decrease in inhibitory interneurons

Table 1. Agents that produce focal epilepsy and their mechanisms of action.

Modified from Dichter and Ayala, 1987.

8

There may be a decrease in inhibition, potentiation of excitation, increase in the space constant

of the dendrites on the postsynaptic neuron, potentiation of neuromodulator release, and/or

supplementation by other voltage-dependent receptors such as N-methyl-D-aspartate receptors

(NMDARs), the slowly inactivating Na+ or Ca

2+ currents, or the large transient Ca

2+ current

(Yamamoto et al. 1980; Connors et al. 1982; Stafstrom et al. 1982; Mayer and Westbrook 1985;

Stafstrom et al. 1985; Stanfield et al. 1985; Herron et al. 1986; Derchansky et al. 2008). A viable

proposal is that potentiated excitatory postsynaptic potentials (EPSPs) will summate and

depolarize the neuron, progressively activating other voltage-gated channels which pass

depolarizing currents such as Na+ and Ca

2+ currents. These will further depolarize the neuron

until the high threshold currents such as the NMDA-mediated current, are activated. This intense

activation of depolarizing currents will thus push the cell to DS and fire a burst of action

potentials (Herron et al. 1985).

Another possible mechanism for DS and burst firing is through the inhibition of K+

currents and subsequent activation of slower Ca2+

currents (Johnston et al. 1980; Stafstrom et al.

1984; Stafstrom et al. 1985). This may be possible through the contribution of neuromodulators

which can reduce K+ currents (Benardo and Prince 1982; Benardo and Prince 1982; Benardo and

Prince 1982; Benardo and Prince 1982; Benardo and Prince 1982; Brown et al. 1982), or

redistributions or changes in the channels in the epileptogenic area due to an injury or lesion.

Following the DS, there is typically a hyperpolarization, which seems to be protective

against the development of seizures, since the disappearance of the hyperpolarization period

often signifies the onset of a SLE (Matsumoto and Marsan 1964; Dichter and Spencer 1969;

Ayala et al. 1970). The most prominent mechanism explaining the hyperpolarizations would be

the presence of inhibitory postsynaptic potentials (IPSPs), such as short latency GABA-mediated

9

IPSPs triggered by the opening of Cl- channels, or longer latency IPSPs resulting from the

opening of K+ channels (Alger and Nicoll 1980). However, the specific conductances involved

in the hyperpolarization may largely depend on what cells are affected, and where the

epileptogenic foci is.

Importantly, seizures are not simply due to hyperexcitability, but it is also necessary to

have hypersynchrony (Wong et al. 1986). Positive feedback can come from recurrent synaptic

excitation, as excitatory recurrent collaterals have been found anatomically in the cortex and

hippocampus, and this recurrent excitation may also help in the amplification of the DS

(Lebovitz et al. 1971; MacVicar and Dudek 1980; MacVicar and Dudek 1980; Knowles and

Schwartzkroin 1981). The intense discharges from neurons can also result in changes in the

ionic environment, with K+ becoming elevated extracellularly, which will cause cell swelling and

a decrease in extracellular space, therefore promoting ephaptic interactions and synchrony

among cells (Konnerth and Heinemann 1983; Yaari et al. 1986). One increasingly prominent

mechanism for synchrony exists in the form of electronic coupling via gap junctions, formed by

connexin proteins of adjacent cells forming a direct channel through which electrical signals can

pass (MacVicar and Dudek 1980; Gutnick and Prince 1981). In fact, it has been shown that gap

junctional blockers can actually block SLEs in seizure models, thus emphasizing the significance

of electrotonic coupling and synchrony in the development of seizure events (Jahromi et al.

2002).

Summarily, the cellular mechanisms involved in epilepsy are complicated and diverse.

There does not seem to be a single unifying theory for all different forms of epilepsies, but on the

most basic level, cellularly, seizures occur as a result of hyperexcitation and hypersynchrony of a

population of neurons in the epileptic focus.

10

1.5 Anticonvulsant Treatment Options

Thus far, no cure for epilepsy exists, and treatments focus on controlling the seizures as a

symptom (Meldrum 2002). The most common treatment is through drug therapy, although there

are some patients who require alternative therapy due to pharmacoresistance. Most antiepileptic

drugs (AEDs) have their effect through modulating or antagonizing channels, in essence,

balancing the hyperexcitable malfunction with another malfunction, and as such, many AEDs

have significant side effects (Meldrum 2002). AEDs can be classified according to their

mechanism of action, and there are five general classifications: ion channel modulators,

enhancers of synaptic inhibition, depressors of synaptic excitation, inducers of cerebral acidosis,

and AEDs with an unknown mechanism of action (Somjen 2004). AEDs often have multiple

actions, and if synergistic, this can be beneficial, but it can also be detrimental if the drug action

affects other channels and other areas of the brain besides the target focus.

AEDs that are ion channel modulators can work on different channels. AEDs acting on

the voltage-dependent Na+ channels (VDSCs) include phenytoin, lamotrigine, valproate,

carbamazepine, and topiramate (Somjen 2004). These drugs act as antagonists of VDSCs, the

primary target being the inactivation of the transient Na+ current (INa,T), and the main effect is the

limiting of sustained high frequency firing that occurs during SLEs (Macdonald and Kelly 1995).

Some AEDs also inhibit the persistent Na+ current (INa,P), which enhances their usefulness in

blocking seizure activity since the INa,P plays a significant role in the repolarization of the cells

after each action potential. Recovery from firing is also slowed and the steady-state inactivation

(h∞) is shifted to a more negative voltage. The main advantage of AEDs that antagonize VDSCs

is that therapeutic dose levels can curb excessive high frequency firing without affecting the

threshold for normal action potentials, so baseline levels of excitability are intact.

11

Some AEDs such as ethosuximide and the methadiones work on voltage-dependent Ca2+

channels (VDCCs). They inhibit low voltage-activated (LVA) Ca2+

currents, and this seems to

inhibit the thalamocortical burst complexes which are characteristic of absence epilepsy (Zhang

et al. 1996). A newer drug, retigabine, also affects ion channels, but it acts by increasing K+

conductance, thus hyperpolarizing neurons and reducing hyperexcitability (Armand et al. 2000).

Beyond acting on ion channels, some AEDs can have an effect by modifying synaptic

transmission. Many AEDs, including phenobarbital, benzodiazepines, gabapentin, felbamate,

topiramate, and also valproate, work by enhancing GABAergic inhibition (Macdonald and Kelly

1995; Macdonald and Greenfield 1997). GABAergic inhibition is enhanced by these drugs

either by inhibiting the reuptake of GABA so the effect of GABA is increased and prolonged at

the synapses, or by increasing the open time or the opening frequency of GABAA receptors.

Other drugs can modify synaptic transmission by antagonizing NMDA receptors, thereby

decreasing levels of excitation (Bialer et al. 2001).

Certain AEDs such as acetazolamide can act primarily through an inhibition of carbonic

anhydrase, and the effect of this is an accumulation of CO2 and corresponding cerebral acidosis

(Heuser et al. 1975; Reiss and Oles 1996). It has been shown that acidosis, whether by a drug

such as acetazolamide or inhaled CO2, can depress activity and result in a depression in cerebral

excitability (Tombaugh and Somjen 1996; Kaila and Ransom 1998). Some other compounds

such as amiloride can cause intracellular acidification by blocking transmembrane acid extrusion,

and this has been shown to successfully ameliorate spontaneous bursting in some hippocampal

seizure models (Bonnet et al. 2000).

An alternative to drug treatment is epilepsy surgery, however, the conditions for a viable

surgical resection are stringent (Awad et al. 1991; Boon et al. 1991). The epileptic focus must be

12

narrowed down to a specific area of the brain, and the patient must also be untreatable with

anticonvulsant medication. The surgery itself must also not involve areas which will result in a

significant neurological deficit if resected (Kwan and Brodie 2000). It is generally understood

that resection of the brain area will disrupt the abnormal activity from propagating from the area

of onset to surrounding regions. Also if the epileptic focus is actually removed, then perhaps the

hyperexcitability itself would no longer pose a problem for the patient.

Another more recent form of treatment involves electrical stimulation. Deep brain

stimulation (DBS) involves the implantation of an electrical stimulator into the brain (Awan et

al. 2009). It is now being used as a treatment paradigm for many neurological disorders, but the

precise mechanisms which underlie this treatment are still unclear. Stimulation can be directly at

the epileptogenic zone (direct control), or indirectly through an anatomical relay of the cortico-

subcortical networks (remote control) since several other structures in the cortico-subcortical

network play supporting roles in epileptic behaviour (Pollo and Villemure 2007).

The first study using DBS for epilepsy involved cerebellar subdural stimulation and there

was a reduction in seizure frequency observed (Cooper et al. 1973). Since then, various other

structures have been targeted including the anterior thalamus, centromedian thalamic nucleus,

caudate nucleus, mamillary body, subthalamic nucleus, and the amygdalohippocampal complex

(see review in (Pollo and Villemure 2007). Experiments in vitro have been done to examine

possible mechanisms for stimulation impinging on seizure activity. In picrotoxin-induced

epileptiform activity in the hippocampus, it has been shown that stimulation can eliminate the

epileptiform discharges. The mechanism is not clear cut, however, since the effects of

stimulation change depending on the frequency and duration of stimulation (Pollo and Villemure

2007). The effect also depends highly on the type of epilepsy, as well as the location of

13

stimulation. Stimulation at remote sites such as the mammillothalamic tract has been shown to

modulate seizure activity by increasing the threshold for producing seizures (Mirski and Fisher

1994). What is more confusing is that in the kainic acid rat model, direct stimulation to the

hippocampus reduces interictal spiking while leaving ictal activity intact (Bragin et al. 2002).

Similar to deep brain stimulation, vagal nerve stimulation also has an effect in some epilepsy

models such as pentylenetetrazol (PTZ)-induced seizures (Woodbury and Woodbury 1990), but

not others, such as the genetic absence epilepsy rats (GAERs) (Dedeurwaerdere et al. 2004).

Ultimately, the mechanisms behind how electrical stimulation can treat seizures are still

largely unknown, and in some cases, stimulation may actually provoke, as opposed to inhibit,

seizures (Pollo and Villemure 2007). There are few main hypotheses for how electrical

stimulation can work. The first is the electrical hypothesis, which suggests that neurons are

inactivated because stimulation induces a large increase in K+, thus shutting down neuronal

activity through a depolarization block where Na+ channels are too depolarized to fire. The

second hypothesis is the neurochemical hypothesis, which proposes that stimulation may

preferentially trigger the release of inhibitory neurotransmitters, thus blocking seizure activity.

Another possibility is that high frequency stimulation may deplete excitatory neurotransmitter

pools and result in a termination of hyperexcitability.

14

2 Potassium and Epilepsy

2.1 Introduction

In order for the brain to function properly, there must be a strict and regulated balance of

ionic concentrations. There are many mechanisms in place to regulate ion concentrations intra-

and extracellularly. Furthermore, the blood brain barrier (BBB) acts as a safeguard to protect the

brain from fluctuations in the blood (Davson and Segal 1996). This chapter will first provide

background information regarding basic membrane physiology and the role of potassium under

standard conditions. Then the focus of the chapter will be on the specific role of potassium and

seizures.

2.2 Basic Membrane Physiology

The membrane potential of a cell is the voltage difference between the interior and

exterior of the cell. The membrane potential (Vm) relation to specific ion concentrations and

permeabilities can be generally expressed through the Goldman-Hodgkin-Katz (GHK) equation:

Where R is the gas constant, T is the absolute temperature, F is Faraday`s number, and P is the

permeability of each ion. This equation leaves out other ions that contribute such as Ca2+

, and

for more accurate approximations, these would have to be taken into account as well. However,

the GHK equation has shown to be adequate and is widely accepted and used for computer

simulations and data analysis.

15

The resting Vm of neurons is a result of the selective permeability of the cell membranes for

specific ions, and it typically rests at approximately -60 to -75 mV, depending on the type of

neuron (Somjen 2004). This is a result of K+ being far more permeable than Na

+ at rest, and

since [K+]e < [K

+]i, there is a net gradient of K

+ moving outward, thereby pushing the cell closer

to the equilibrium potential of K+ (~-80 mV). To maintain the electrochemical gradient between

the interior and exterior of the cells, active transport processes are required, and the most

important pump is the Na+/K

+ ATPase, which uses energy from ATP to pump 3 Na

+ out and 2

K+ into the cell.

When there is excitatory synaptic input to the dendrites, EPSPs summate in the soma and

when the membrane potential is depolarized beyond threshold (~-45 mV), Na+ channels are

opened and the permeability of Na+ becomes much greater than that of potassium (Hodgkin et al.

1952). The transient sodium current INa,T then depolarizes the cell closer to the equilibrium

potential of Na+ (~ +40 mV) resulting in sharp depolarization that is the action potential. The

action potential is very brief however, due to the Na+ channels becoming inactivated from

depolarization. Following the action potential, there is typically a hyperpolarization due to the

slower kinetics of the voltage-dependent K+ channels. This allows K

+ currents, particularly the

delayed rectifier and smaller A currents (IK,DR and IK,A respectively) to drive the Vm to

hyperpolarized levels. When these K+ channels close, the cell repolarizes back to its resting Vm,

aided by the small inward persistent Na+ current (INa,P). With the cell repolarized, the Na

+

channels are un-inactivated and the refractory period ends.

It is clear then, that each time a cell fires, K+ is released and the ratio of [K

+]e / [K

+]i

increases. This elevation in extracellular K+ can have major consequences on cellular function,

16

and as a result, there are certain safeguards in place for buffering K+ elevations. During low

frequency firing, the Na+/K

+ ATPase is sufficient to return released K

+ back to the cell, and it

ensures that the elevation in K+ is only minimal and also transient (Connors et al. 1979). In

situations of focal high-frequency firing, the elevation in K+ will be very concentrated in a small

area, so following the concentration gradient, the K+ can diffuse to neighboring areas

temporarily, and as a result, after firing abates and the Na+/K

+ pump works to pump K

+ back into

cells, there is a noticeable postexcitation undershoot prior to K+ returning to baseline levels

(Heinemann and Lux 1975). Lastly and most powerfully, glial cells, particularly astrocytes,

work both passively and actively to buffer elevations in K+ by taking the excess K

+ and

transporting it through the glial network through gap junctions, thereby dispersing the K+

(Ballanyi et al. 1987).

2.3 The Role of Potassium in Seizure Activity

The connection between K+ and seizure activity has been studied for many decades. It

was a widely debated issue whether the elevated K+ is a cause or an effect of seizures.

The potassium hypothesis of seizure generation proposes that there may be a positive

feedback loop involving hyperexcited neurons and excess K+ release and this would thus,

reinforce the excitation and encourage further seizure activity until depolarization reached levels

at which VDSCs would be kept in an inactivated state and be unable to fire until repolarized. At

this level, seizures would terminate and K+ levels would return to normal (Figure 1) (Green

1964; Fertziger and Ranck 1970). Feldberg and Sherwood showed in cats that KCl injections

into the lateral ventricle would induce epileptic convulsions in the cats (Feldberg and Sherwood

1957).

17

Figure 1. The Potassium Hypothesis of Seizure Generation.

This hypothesis proposes that there is a positive feedback between hyperexcited neurons

releasing K+ and the elevated K

+ further exciting the neurons until there is a depolarization

block, at which point both K+ elevation and neuronal activity ceases, marking the

termination of SLEs. Modified from Fertziger and Ranck, 1970.

18

It was later demonstrated that if artificial CSF (aCSF) with elevated K+ was perfused through the

lateral ventricle, seizure activity would be seen in the hippocampal region (Zuckermann and

Glaser 1968; Zuckermann and Glaser 1968; Zuckermann and Glaser 1968). Most labs have [K+]

of ~2.5-3.5 mM in their aCSF solution, and it has been found that even slight elevations up to 5

mM K+, can greatly facilitate paroxysmal firing triggered by repetitive electrical stimulation to

afferent pathways (Somjen et al. 1985). With further elevations in K+, up to ~8 mM,

spontaneous activity can be triggered in the form of interictal discharges or prolonged SLEs

(Rutecki et al. 1985; Korn et al. 1987; Traynelis and Dingledine 1988; Jensen and Yaari 1997).

There are many reasons why elevated K+ may promote SLEs, some of which are summarized in

Figure 2 (Traynelis and Dingledine 1988). Elevated [K+]e depolarizes cells, moving the potential

closer to firing threshold and increasing NMDAR activation. It causes the cell to swell, which

decreases the extracellular space and increases ephaptic coupling, which aids in synchronization

of the network. It also depolarizes presynaptic terminals and cause activation of VDCCs and

inappropriate transmitter release (Hablitz and Lundervold 1981). GABAA-mediated synaptic

inhibition may also be rendered less effective with elevated K+, as it has been shown that this

elevation can force the uptake of Cl-, possibly through activation of an inward KCl pump, and

shift the reversal potential of GABAergic IPSPs to more depolarized levels (Korn et al. 1987;

Chamberlin and Dingledine 1988). A further possibility is that K+ can have secondary effects

through the potentiation of other currents such as the INa,P (Somjen and Muller 2000).

The potassium hypothesis has been widely debated for many years, but by now, it is

largely rejected due to observations made using K+-sensitive electrodes (Futamachi et al. 1974;

Lux 1974; Pedley et al. 1976; Benninger et al. 1980; Somjen et al. 1986).

19

Figure 2. Proposed Effects by which Elevated K+ Triggers SLEs.

The elevation in [K+] can have many effects, which all work together in a positive feedback

loop. These effects are then proposed to promote the initiation of SLEs.

Modified from Traynelis and Dingledine, 1988.

20

The potassium hypothesis predicts that there is a threshold [K+]e for seizure initiation, but no set

threshold has been found (Lux 1974). The elevation in K+ also seems to lag behind seizures as

opposed to leading them. This suggests that the elevated K+ may be a consequence of the

seizures as opposed to a cause. Another argument against the hypothesis is that electrical

stimulation in healthy tissue can increase K+ levels to that seen during seizures without triggering

seizure activity. A possible defense for the potassium hypothesis is that K+-sensitive electrodes,

since they are double-bored, may create a liquid-filled cavity that dilutes or delays the K+

measurements made (Somjen 2004). Furthermore, it has been seen that electrical stimulation can

elevate K+ to a ceiling level, but upon initiation of SLEs, the K

+ further increases to an even

higher ceiling level (Somjen and Giacchino 1985; Stringer and Lothman 1989; Stringer et al.

1989).

In any case, it is without dispute that the elevation in K+, regardless of whether it is cause

or consequence, will have a major impact on neuronal function. It has been proposed that

interictal discharges may not be caused by elevated K+, but the elevated K

+ resulting from these

discharges may push the activity from interictal to ictal (Dichter et al. 1972; Jensen and Yaari

1997; Borck and Jefferys 1999). Preceding ictal onset, it was observed that K+ would increase

and the SLE would occur from a set threshold K+

of ~7.5 mM. During the SLE, K+ levels would

reach the documented ceiling of 12 mM. The K+ may be contributing to the interictal to ictal

transition not only by increasing the firing frequency of spontaneously active cells, but also by

recruiting otherwise inactive cells (Cohen and Miles 2000).

21

3 The Ih Current

3.1 Introduction

The hyperpolarization-activated (Ih) current is a voltage-activated inward cationic current

that was first studied in the heart (Noma and Irisawa 1976; Brown et al. 1979; DiFrancesco

1986; DiFrancesco et al. 1986). It was first termed the “funny” current (If) because of its

activation by hyperpolarization as opposed to the characteristic depolarization. This current is

known as a pacemaker current which allows for the sino-atrial cells to maintain spontaneous

activity. It works by activating during the repolarization after action potentials, thus helping

depolarize the cell again and, along with VDCCs, pushing the cells to generate an action

potential again. Since then, this current has been found in neurons, including in the hippocampus

(Halliwell and Adams 1982). First termed “queer” current (Iq) in central neurons, both the If and

Iq were later grouped and termed as Ih to prevent confusion. For a review on the many factors

that regulate the Ih current, see review by Wahl-Schott and Biel, 2009, but for the purposes of

this thesis, these factors will not be discussed in detail (Wahl-Schott and Biel 2009). This

chapter will focus on the characteristics of the Ih, its functional properties, and its role in

seizures.

3.2 HCN Channel Structure

As reviewed by Wahl-Schott and Biel, 2009, the Ih flows through hyperpolarization-

activated and cyclic-nucleotide-gated (HCN) channels (Wahl-Schott and Biel 2009). The HCN

channels are part of a superfamily of voltage-gated pore loop channels, and are encoded by four

genes: HCN1-4.

22

HCN channels are tetrameric and, in mammals, there are four different homotetramers

each with distinct biophysical properties (Wahl-Schott and Biel 2009). In vivo, heterotetramers

can form and, as such, there are more HCN channel types (Ulens and Tytgat 2001). HCN

channels have two modules which cooperate allosterically during channel activation: the

transmembrane core and the cytosolic C-terminal domain. The transmembrane core is composed

of six α-helices (S1-S6) and an ion-conducting pore loop between the S5-S6 segment. The

voltage sensor, as is characteristic in all voltage-dependent members of this family, is found on

the positively charged S4 segment (Yu and Catterall 2004). As opposed to conventional

depolarization-activated channels, however, it is inward movement of the S4 segment that

triggers opening of HCN channels (Mannikko et al. 2002). The selectivity filter of HCN

channels have with a glycine-tyrosine-glycine (GYG) motif characteristic of K+ selective

channels, and as such, HCN channels are most permeable to K+. They are also highly sensitive

to K+ fluctuations, and it has been shown that even slight elevations in K

+ can result in a very

marked increase in Ih conductance as well as a depolarizing shift in reversal potential (Spain et

al. 1987; Funahashi et al. 2003; Shin and Carlen 2008). However, these channels are also

permeable to Na+ and on a lesser level, Ca

2+ (Pape 1996; Yu et al. 2004; Yu et al. 2007). The

reversal potential lies between -25m mV and -40 mV and the cations driving the inward current

depend on the membrane potential and the corresponding reversal potentials of the cations

(Robinson and Siegelbaum 2003).

The cytosolic C-terminal domain is important largely because of the cyclic nucleotide-

binding domain (CNBD). Unlike cyclic nucleotide-gated (CNG) channels, HCN channel

opening is largely voltage-dependent and CNs act as modulators (DiFrancesco and Tortora

1991). Another difference is that HCN channels are modulated by CNs in a phosphorylation-

23

independent manner. The HCN channels can be modulated by both cyclic adenosine

monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), though the affinity for

cAMP is reportedly ~10-100-fold higher than for cGMP (Ludwig et al. 1998). When CN levels

are elevated, typically by hormones or neurotransmitters, there is a shift in the half-maximal

activation (V1/2) from ~ -70 mV to more positive values (DiFrancesco and Tortora 1991). The

opening kinetics are also accelerated by shifting the voltage dependence of activation. This may

be because when CNs are not bound, the β-roll subdomain of the CNBD inhibits channel

activation and shifts it to more negative potentials, but when CNs are bound, this inhibition may

be relieved, thereby shifting the gating to more positive values and accelerating opening kinetics

(Wainger et al. 2001). Beyond being a modulatory domain, it has been proposed that the CNBD

may also be important for normal cell surface expression of the protein subunits (Proenza et al.

2002).

3.3 HCN Channel Expression

As the functional properties of each of the subunits differ, so do their levels of expression

based on the region, the cell type, and the age (Ludwig et al. 1998; Moosmang et al. 1999;

Bender et al. 2001; Moosmang et al. 2001). In the hippocampus, through mRNA expression

analysis, it has been found that all of the HCN subunits are present, although HCN1 and HCN2

seem to predominate. Bender et al., 2001 found that during development, HCN1 was largely

present in the hippocampal pyramidal cell layers for both CA1 and CA3 regions (Bender et al.

2001). The presence of HCN1 in the CA3 principal cells seems to be important for the gamma

frequency oscillations observed early in development (Palva et al. 2000). By the third postnatal

week, HCN1 expression is absent in the CA3 principal cells, and instead expression becomes

24

focused in the CA1 principal cells, as well as in the interneurons in both CA1 and CA3

pyramidal layers, specifically the basket and chandelier cells (Bender et al. 2001). It is in this

third postnatal week, the hippocampus is seen to undergo a functional remodeling which reduces

the excitability of the adult brain, and the corresponding absence of HCN1 in the CA3

pyramidals may contribute to this (Smith et al. 1995). The basket cell and chandelier cell

interneurons of the hippocampus display a high frequency spiking pattern, and the fast activating

HCN1 channels expressed on these interneurons may play a role in this (van Hooft et al. 2000).

It is proposed that HCN1 in these interneurons may participate in frequency regulation of the

tonic inhibitory input to the principal cells, similar to its role in cerebellar basket cells (Saitow

and Konishi 2000).

HCN4 generally follows the expression pattern of HCN2, which is homogenously

expressed throughout the principal cell layers and expressed more copiously in the interneuronal

populations in the stratum oriens and alveus (O/A) (Bender et al. 2001). O/A interneurons have

been shown to exhibit rhythmic firing activity, and this has been attributed partly to Ih

(Maccaferri and McBain 1996). It is likely that the slow-activating HCN2 and HCN4 contribute

to this rhythmic firing activity and that this activity may provide feedback inhibition to

pyramidal cells, thereby modulating synchronous activity in the pyramidal layers (Katona et al.

1999). HCN3 is not significantly expressed in the hippocampus (Bender et al. 2001).

The density, properties, and subcellular distribution of HCN channels are subject to

activity-dependent regulation (Dyhrfjeld-Johnsen et al. 2009). In the hippocampal CA1

pyramidal cells, the increasing distribution along the apical dendrites is dependent on excitatory

input from the entorhinal cortex (Shin and Chetkovich 2007). It has been proposed that TPR-

containing Rab8b interacting chaperone proteins (TRIP8b) may co-localize with HCN1 subunits

25

in CA1 pyramidal neurons and work to quickly insert or remove HCN channel proteins in an

activity-dependent manner (Santoro et al. 2004).

3.4 Functional Properties of Ih

It is known that Ih contributes to two major physiological roles: it helps set the RMP, and

it helps provide negative feedback for membrane potential stabilization (Pape 1996; Doan and

Kunze 1999; Wahl-Schott and Biel 2009). First, Ih is constitutively open at rest and it helps set

where the RMP sits by passing a depolarizing noninactivating inward current. Ih also decreases

the membrane resistance (Rm), which functions to stabilize the Vm, suppress low frequency

fluctuations, and attenuate the amplitude of incoming EPSPs (Nolan et al. 2007). In the presence

of Ih, the kinetic filtering of EPSPs is also affected, so EPSP rise and decay time is accelerated.

Secondly, Ih provides negative feedback. With the reversal potential of Ih sitting at around -25

mV to -40 mV, depending on the cell-type and region, when the Vm is at more hyperpolarized

voltages, the Ih provides a depolarizing inward current. On the other hand, at more depolarized

potentials, the Ih shuts off and the withdrawal of this tonic depolarizing inward current results in

a hyperpolarization.

Ih has been shown to be involved in many aspects of physiology, including working

memory, hippocampal LTP, motor learning, and pacemaking oscillations (Wahl-Schott and Biel

2009). Ih has also been found to be present in the presynaptic terminals of the crustacean

neuromuscular junction (Beaumont and Zucker 2000), avian ciliary ganglion (Fletcher and

Chiappinelli 1992), cerebellar basket cells and the calyx of Held (Southan et al. 2000), but what

function presynaptic Ih has is still debated. Some have suggested that presynaptic Ih may be

involved in hippocampal long term potentiation (LTP) (Mellor et al. 2002), but other studies in

26

the same system show contradictory results (Chevaleyre and Castillo 2002). Importantly, Ih is

involved in signal processing since it plays a major role in dendritic integration (Magee 1999;

Magee and Carruth 1999). Passive cable theory will suggest that distal synaptic input should

generate slower EPSPs in the soma as compared to proximal input. However, dendrites are not

voltage-independent and are not passive since they express a multitude of voltage-dependent

channels, including HCN channels. As mentioned earlier then, experiments have since shown

that Ih decreases the Rm and increases local membrane conductance, thereby accelerating the

speed of EPSP rise and decay (Magee 1999; Magee 2000; Nolan et al. 2007).

3.5 Ih and Epilepsy

The role of Ih in epilepsy has been examined in many seizure models, including the

perinatal seizure-inducing hypoxia model (Zhang et al. 2006), kainate model (Shah et al. 2004),

pilocarpine model (Jung et al. 2007; Shin et al. 2008; Marcelin et al. 2009), fluid percussion

model (Howard et al. 2007), and the febrile seizure model (Chen et al. 2001; Dyhrfjeld-Johnsen

et al. 2008). However, the role of Ih is still highly debated, as some studies have shown

increased Ih promoting seizures, while others shown an attenuation of seizures with increased Ih.

In the pilocarpine, kainate, and perinatal hypoxia models, there was a reduction in Ih, and it

seemed that the resulting hyperexcitability was due to an increase in neuronal input resistance

(Shah et al. 2004; Zhang et al. 2006; Jung et al. 2007; Shin et al. 2008; Marcelin et al. 2009). A

recent paper also showed that in human epileptogenic neocortex, a deficit of HCN1 subunits may

be responsible for the increase in excitability and probability of seizure activity (Wierschke et al.

2009). On the other hand, a recent study on febrile seizures showed an increase in Ih current,

along with a depolarized V1/2 (Dyhrfjeld-Johnsen et al. 2008). In studies of synaptic plasticity, it

has been shown that LTP induction can increase HCN channel protein synthesis and decrease

27

excitability (Fan et al. 2005) while long term depression (LTD) induction downregulates Ih and

increases neuronal excitability (Brager and Johnston 2007).

In examining the kainate model for temporal lobe epilepsy, the downregulation of HCN1

channels has been suggested to be linked to a corresponding seizure-induced upregulation of

neuron-restrictive silencing factor (NRSF), which restricts transcription of HCN1 (Dyhrfjeld-

Johnsen et al. 2009). Another possibility is that the interaction between the TRIP8b chaperone

protein and the HCN1 subunits is disrupted, so there is a channelopathic mislocation of h-

channels, leading to the observed downregulation (Santoro et al. 2004). In the febrile seizure

model, there is also a downregulation of HCN1 subunit expression, but the Ih is increased (Chen

et al. 2001; Dyhrfjeld-Johnsen et al. 2008). It is suggested that this may be a result of the

seizures causing a glycosylation of HCN1 subunits, prompting the formation of HCN1/HCN2

heteromeric channels which have different properties than the former HCN1 homomeric

channels (Brewster et al. 2005). These heteromeric channels may then explain the increase in Ih

and the changes in the properties of Ih, including the more depolarized activation voltage.

In summary, there is a delicate balance between Ih-induced depolarization, which brings

the Vm closer to the threshold of firing, making it more excitable, and an Ih-induced decrease in

input resistance, which decreases summation of EPSPs and making the cells less excitable.

Different models may affect the interplay of these parameters in different ways. Adding to the

complication is that Ih exerts influence through the interaction of other voltage-gated channels.

It has been proposed that, in the febrile seizure model, part of the increased excitability from the

upregulated Ih may also be due to a concomittent downregulation of a persistent K+ current

(Dyhrfjeld-Johnsen et al. 2008). Ih may also interact with inhibitory inputs by causing a post-

inhibitory rebound depolarization (Kitayama et al. 2003).

28

4 Research Rationale, Objectives, and Hypotheses

Research Rationale and Objectives:

It is well established that there is a strong connection between [K+]e and seizures,

however, the role of [K+]e in seizure cessation has not been thoroughly studied. It has been

postulated that elevated [K+]e may stop seizures through a depolarization block, whereby VDSCs

are inactivated (Green 1964; Fertziger and Ranck 1970), however, this postulate has not been

thoroughly characterized and the role of ionic currents affected by elevations in [K+]e thus far

have not been studied in detail.

The main objective of this thesis is to expand upon the hypothesis that [K+]e may become

anticonvulsant, and to examine the cellular mechanisms involved in this switch in the activity of

[K+]e. Here I characterize the range at which [K

+]e switches from being pro- to anticonvulsant in

the in vitro low Mg2+

/high K+ and tetanization models, and I also examine whether the anti-

convulsant properties are simply a result of Goldman-Hodgkin-Katz (GHK)-mediated

depolarization, or if Ih is involved in this switch in the activity of [K+]e in epilepsy.

Hypotheses:

My hypothesis is threefold:

(1) That in epileptic tissue, further elevations in [K+]e can switch its activity from pro-

convulsant to anti-convulsant

(2) This switch in the action of K+ cannot be accounted for strictly by Goldman-Hodgkin-

Katz (GHK)-mediated depolarization

(3) That there is an ionic current, Ih, responsible for the switch from pro- to anti-convulsant

in K+

29

5 Materials and Methods

Animals and dissecting protocol:

Experiments were conducted on juvenile male Sprague-Dawley rats post natal day 17-21

in compliance with the guidelines of the University Health Network Animal Care Committee.

Rats were anaesthetized with halothane and quickly decapitated. The brain was rapidly removed

and transferred to ice cold artificial cerebral spinal fluid (aCSF) dissecting solution containing, in

mM: 75 sucrose, 87 NaCl, 25 NaHCO3, 7 MgCl2, 1.25 NaH2PO4, 2.5 KCl, 25 D-Glucose, 0.5

CaCl2, pH adjusted to 7.4 with 95% O2/5% CO2 (carbogen). Hippocampal slices were prepared

as follows: the brain was hemisected along the midsagittal line and the cerebellum and the

forebrain were removed. The dorsal cortex was cut parallel to the longitudinal axis and the brain

was secured, ventral side up, into a vibratome (Leica VT1200, Leica Microsystems, Heidelberg,

Germany) and 400 µm horizontal oblique slices were obtained. Afterwards, slices were left to

incubate in aCSF heated to 30°C for 1 h to allow for tissue stabilization and recovery before

recording. This incubation solution or aCSF contained in mM: 125 NaCl, 2.5 KCl, 25 D-

Glucose, 25 NaHCO3, 1.25 NaH2PO4, 1 MgCl2, 1 CaCl2. It was continually aerated with 95%

O2/5% CO2.

Field recording protocol:

Transverse brain slices were placed in an interface-type chamber for recording

extracellular field responses and perfused with aCSF at 2-3 mL/min, aerated and pH balanced

with 95% O2/ 5% CO2 at 34.0°C. While the slice was present in the interface chamber,

humidified oxygen was constantly gassed on top of the slice. Schaffer collaterals were

30

stimulated by a bipolar electrode (125 µm in diameter, platinum/iridium biconcentric, FHC, ME,

USA) placed in the stratum radiatum of the CA1. Orthodromic extracellular responses were

recorded largely from CA1 pyramidal neurons with borosilicate glass pipettes filled with aCSF

with a tip resistance of 2-4 MΩ in the bridge balance mode. A few recordings were also

performed with the recording electrode in the stratum radiation of the CA1 to examine the

presynaptic volley. Constant current stimulating square pulses of 100 µs duration were delivered

via a constant current isolation unit connected to a Grass S88 stimulator (Grass Instruments,

Quincy, MA, USA). Responses were sampled at 2 KHz and low pass filtered at 5 KHz with an

Axoclamp 2A amplifier (Molecular Devices, Sunnyvale, CA, USA). Various stimulating pulses

were applied to evoke subthreshold and maximal suprathreshold potentials to construct an input-

output curve. The stimulus strength required to elicit the maximal field excitatory postsynaptic

potential (fEPSP) and population spike was determined from the input-output curve. This

stimulus strength was typically ~3 mA and was kept constant throughout the experiment during

the various treatment paradigms. To quantify field potentials in the CA1 pyramidal layer, the

most positive potential was subtracted from baseline while population spikes were measured

from the leading positive edge of the population spike to the most negative potential. Data

acquisition and analyses were performed using pClamp 9.2 (Molecular devices, Forest City, CA,

USA). For gap-free recordings, an offline Bessel (8-pole) high pass filter set at 0.2 Hz cutoff

was applied to filter out background fluctuations.

Low Mg2+

/5 mM K+ seizure induction protocol:

SLEs were induced by perfusing slices for 30 min. with low Mg2+

(0.25 mM)/high K+ (5

mM) after an initial 5 minute aCSF baseline. Input and ouput (I/O) relations were recorded prior

31

to, and after perfusion with different solutions, where the [K+] was altered in the range of 6 mM

to 12.5 mM, while maintaining low Mg2+

.

Tetanic stimulation-induced afterdischarge protocol:

Primary afterdischarges (PADs) were evoked in CA1 pyramidal neurons by tetanic

stimulation of Schaffer collaterals with repetitive pulses of 100-µs duration at 100 Hz for 2

seconds (Rafiq et al. 1993; Jahromi et al. 2002). Tetanization of slices was repeated once every

10 minutes, 6-10 times, depending on when robust PADs are evoked. Tetanization experiments

carried out in 2.5 mM K+ were performed separately. After 6-10 tetanizations in 5 mM K

+, K

+

was further elevated to 7.5 mM K+ to see the effect of the further elevation of K

+ on the PADs.

Treatment with Ih Blocker

In the low Mg2+

/5 mM K+ seizure induction model, after SLEs had been reliably induced, 50 µM

of the hyperpolarization-activated nonspecific cationic current (Ih) blocker, 4-ethylphenylamino-

1,2-dimethyl-6-methylaminopyrimidinium chloride (ZD7288) was included in the perfusion for

concentrations at 5 mM K+ and 7.5 mM K

+. ZD7288 is well-used for blockade of Ih and its

effect has also been studied with respect to seizure activity. Dose-response tests have been

shown in literature, from a dose of 10 µM to 100 µM, and at the higher end of these

concentrations, it seems to block synaptic transmission (Inaba et al. 2006). 50 µM has been

previously tested in our lab, and has been shown to effectively block Ih, while synaptic

transmission is still preserved (Shin and Carlen 2008), so 50 µM was used as our testing dose.

32

Drugs:

All chemicals were purchased from Sigma-Aldrich (Burlington, ON, Canada), except

ZD7288, which was purchased from Tocris Bioscience (Ellisville, MO).

Statistical Analysis:

For analysis of data in the low Mg2+

/high K+ model, all evoked field potentials and

population spikes in the various treatment groups were normalized to values obtained in aCSF.

To analyze seizure amplitude and duration during perfusion of low Mg2+

/high K+ (in the range of

6 mM to 12.5 mM), all SLE amplitudes and durations were normalized to values recorded in low

Mg2+

/5 mM K+. Prior to running any statistical test, normalized data were converted to a normal

distribution using a root arcsine conversion. For analysis of data in the tetanization model, mean

amplitudes of evoked population spikes and fEPSPs, as well as mean amplitudes and durations

of PADs were compared. Significance levels were determined at P<0.05. Each sample size (n)

equated to a single slice. All data are expressed as mean ± SEM. Statistical analysis was carried

out using one-way analysis of variance (One way-ANOVA) with a Dunnett’s post-hoc (against

control) analysis using Sigmastat 3.5 software program (Systat Software Inc, San Jose, CA,

USA).

33

6 Results

6.1 The Effect of Elevating [K+]e to Ceiling Levels Observed During SLEs

To induce SLEs, brain slices containing the hippocampal formation were bath perfused

with low Mg2+

/high K+ for 30 minutes. For this study, SLEs were defined as spontaneous high

amplitude, high frequency events lasting a minimum duration of 5 seconds. Field recordings

were obtained from the CA1 hippocampal pyramidal layer during aCSF perfusion. No SLEs

were observed during this treatment (Figure 3C). Single stimulus-evoked field potentials

produced stereotypical responses in the pyramidal layer – a population spike of 1.62 ± 0.15 mV

(n=23) preceding a fEPSP of 1.02 ± 0.10 mV (n=25) (Figure 3A). During a 30 minute perfusion

of low Mg2+

/5 mM K+, neither the population spike or fEPSP amplitudes (84.9 ± 17.8% and

117.6 ± 15.8% respectively) were significantly different from the control levels (Figures 3B,

6A,B). With low Mg2+

/5 mM K+ perfusion, SLEs were observed to spontaneously occur at 7-10

minute intervals (Figures 3D,6C).

After spontaneous SLEs were reliably recurring, the [K+]e was elevated in the bath up to

ceiling levels (10-12 mM) typically observed with seizures. At 12.5 mM [K+]e, the highest level

of [K+]e tested, single stimuli were unable to evoke population spikes or fEPSPs (n=4) (Figures

4A, 6A,B), whereas at 10 mM [K+]e (n=3), the evoked population spikes and fEPSPs were

significantly blunted to 31.5 ± 25.8% and 28.0 ± 14.1% of aCSF values respectively (n=3)

(P<0.05) (Figures 4B, 6A,B). After perfusion for 30 mins at both of these [K+]e, SLEs were

completely blocked (Figure 4C,D,6C). At 10 mM [K+]e, there was tonic low amplitude

background activity (Figure 4D). Upon switching back to 5 mM [K+]e, SLEs returned and upon

washout back to aCSF, SLEs stopped and the evoked population spike and fEPSP amplitudes

were not significantly different from the initial values.

34

Figure 3. Evoked responses and spontaneous potentials in aCSF and low Mg2+

/5 mM K+.

The stimulus strength for evoked responses was determined by an input-output curve and

the maximal field and population spike responses are shown for aCSF (A) and in low

Mg2+

/5 mM K+ (B). In low Mg

2+/5 mM K

+, multiple population spikes could be seen,

indicating hyperexcitability. Spontaneous field potentials are shown for aCSF (C) and low

Mg2+

/5 mM K+ (D). There was no observable activity seen during aCSF perfusion, whereas

during low Mg2+

/5 mM K+, spontaneous recurrent SLEs would occur at regular intervals

after the initial SLE. The scale bar (1 mV/20 ms) represents all evoked responses. The

scale bar (1 mV/50 s) represents the spontaneous field potential recordings in aCSF. The

scale bar (2 mV/2 min) represents the spontaneous field potential recordings in low Mg2+

/5

mM K+

35

Figure 4. Evoked responses and spontaneous potentials in low Mg2+

/12.5 mM K+ and low

Mg2+

/10 mM K+.

The stimulus strength for evoked responses was determined by an input-output curve and

the maximal field and population spike responses are shown for low Mg2+

/12.5 mM K+ (A)

and low Mg2+

/10 mM K+ (B). In low Mg

2+/12.5 mM K

+, no population spike or fEPSP

could be measured, and in low Mg2+

/10 mM K+, the evoked population spike and fEPSP is

significantly attenuated (P<0.05). Spontaneous field potentials are shown for low Mg2+

/12.5

mM K+ (C) and low Mg

2+/10 mM K

+ (D). After the transition into low Mg

2+/12.5 mM K

+ is

stabilized, no activity is seen. After stabilization in low Mg2+

/10 mM K+, tonic low

amplitude background activity is seen. (E) shows an expanded trace of this background

activity. The scale bar (1 mV/20 ms) represents all evoked responses. The scale bar (2

mV/100 s) represents all spontaneous field potential recordings not expanded. The scale

bar (1 mV/5 s) represents the expanded trace in (E).

36

6.2 Finding the Anticonvulsant Range of [K+]e in the Low Mg2+/5 mM K+ Model

After spontaneous, recurrent SLEs have been elicited through perfusion of low Mg2+

/5

mM K+, [K

+]e was elevated in the bath to a concentration of 7.5 mM. In this study, in low

Mg2+

/7.5 mM K+, the population spike and fEPSP amplitudes were significantly reduced to 46.3

± 23.4% and 30.3 ± 15.2% of the control aCSF values (n=4) (P<0.05) (Figures 5A,6A,B). SLEs

were also attenuated and replaced with tonic low amplitude background activity, similar to what

was observed in low Mg2+

/10 mM K+ (Figures 5D,6C). At this concentration, some recordings

were also performed in the stratum radiatum to observe the presynaptic volley, and no significant

difference in the amplitude of the synaptic volley was seen (data not shown) (n=4). All other

recordings were performed in the stratum pyramidale, and the presynaptic volley cannot be

clearly discerned from the stimulus artifact.

To narrow down the anticonvulsant range of [K+]e, another set of experiments was carried

out where [K+]e in the low Mg

2+/5 mM K

+ bath was increased to 6 mM. The population spike

and fEPSP amplitudes (66.6 ± 12.5% and 133 ± 66.5% of control aCSF values respectively)

were not significantly changed (n=3) (Figures 5C,6A,B). With low Mg2+

/6 mM K+ perfusion,

SLEs were not attenuated, and the SLEs had similar amplitudes and durations as compared to

SLEs seen in low Mg2+

/5 mM K+ (80.5 ± 10.4% and 106.0 ± 26.0% respectively) (Figures

5G,6C).

Since low Mg2+

/6 mM K+ allowed SLEs to continue unhindered, while low Mg

2+/7.5 mM

K+ attenuated SLEs, we hypothesized that the lower boundary for the anticonvulsant properties

of elevated [K+]e would lie between those [K

+]e values. We increased the [K

+]e in the low

Mg2+

/5 mM K+ solution to 6.5 mM K

+. At this [K

+]e, the amplitude of population spikes and

fEPSPs (94.3 ± 24.8% and 72.3 ± 17.8% respectively) were not significantly different than in

37

control aCSF (Figures 5B,6A,B). At 6.5 mM K+, 50% of the slices (n=3) continued to seize in

the elevated [K+]e (Figure 5E), whereas in the other 50% (n=3), seizures were attenuated and

tonic low amplitude background activity was observed (n=3) (Figure 5F). Of the slices that

seized, the seizure amplitudes and durations (85.8 ± 8.0% and 104.5 ± 16.4% respectively

compared to low Mg2+

/5 mM K+ seizures) were similar to those seen in low Mg

2+/5 mM K

+ but

there was also an increase in interictal discharge between SLEs (Figures 5E,6C).

38

Figure 5. Evoked responses and spontaneous potentials in low Mg2+

/7.5 mM K+, low

Mg2+

/6.5 mM K+, and low Mg

2+/6 mM K

+.

The stimulus strength for evoked responses was determined by an input-output curve and

the maximal field and population spike responses are shown for low Mg2+

/7.5 mM K+ (A),

low Mg2+

/6.5 mM K+ (B) and low Mg

2+/6 mM K

+ (C). In low Mg

2+/7.5 mM K

+, the evoked

population spike and fEPSP was significantly attenuated (P<0.05) whereas there was no

significant change in the population spike and fEPSP amplitudes in low Mg2+

/6.5 mM K+

and low Mg2+

/6 mM K+. Spontaneous field potentials are shown for low Mg

2+/7.5 mM K

+

(D), low Mg2+

/6.5 mM K+ with SLEs (E), low Mg

2+/6.5 mM K

+ when SLEs are attenuated

(F), and low Mg2+

/6 mM K+ (G). For low Mg

2+/7.5 mM K

+ and low Mg

2+/6.5 mM K

+ when

SLEs are attenuated, there is tonic low amplitude activity seen throughout the trace (D&F).

For low Mg2+

/6.5 mM K+ when SLEs persist, there is increased interictal firing between

SLEs (E). In low Mg2+

/6 mM K+, there is no significant difference in the amplitude or

duration of the SLEs (G). The scale bar (1 mV/20 ms) represents all evoked responses.

The scale bar (2 mV/2 min) represents all spontaneous field potential recordings.

39

Figure 6. Normalized group data showing the effects of elevating [K+]e on the population spike and fEPSP amplitude, and the

SLE amplitude and duration in the low Mg2+

/5 mM K+ model.

The dotted line represents the values the data is normalized to. In A and B, it represents the aCSF value and in C it represents

the low Mg2+

/5 mM K+ value. Asterisks denote significant changes against the control reference values (P<0.05). In all

parameters, a significant decrease is found at 7.5, 10, and 12.5 mM [K+]e. In C, the values for 6.5 mM K

+ are based only on the

50% of slices that had seizures.

40

6.3 The Effects of Elevating [K+]e in the Tetanization Model

In control aCSF containing 2.5 mM K+, 100 Hz tetanizations for 2 sec periods every 10

minutes elicited PADs after 6-10 tetanizations. In 2.5 mM K+, the population spikes had an

amplitude of 1.57 ± 0.27 mV and the fEPSPs had an amplitude of 1.0 ± 0.13 mV (n=4) (Figure

7A). After 6-10 tetanizations, the measured PAD amplitude was 0.31 ± 0.07 mV and the PAD

lasted for a duration of 13.93 ± 2.80 seconds (Figure 7D,G,H).

When the same experimental paradigm was run in aCSF with 5 mM K+, the population

spike amplitude (1.26 ± 0.02 mV) and the fEPSP amplitude (1.55 ± 0.29 mV) did not change

significantly from values in control aCSF (n=4) (Figure 7B). However, the PADs evoked by

tetanization in the 5 mM K+ were significantly higher amplitude (0.68 ± 0.13 mV) and longer

duration (24.53 ± 3.47 seconds) (n=4) (P<0.05) compared to PADs evoked in control aCSF

(Figure 7E,G,H). Further experiments were conducted where the [K+]e in the bath was further

elevated to 7.5 mM to study the effect of elevated K+ on tetanization-induced PADs in 5 mM K

+.

This is because the PADs were much more pronounced at 5 mM K+, and in previous papers,

tetanization experiments are carried in aCSF with a baseline of 5 mM K+

(Jahromi et al. 2002).

When the [K+]e in the bath was increased from 5 mM K

+ up to 7.5 mM K

+, there was no

significant difference in the amplitude of population spikes and fEPSPs (Figure 7C). However,

in 7.5 mM K+, PADs could not be evoked by tetanization (n=4) (Figure 7F,G,H). Instead, tonic

background activity resembling interictal spikes was observed throughout the recording. The

tonic interictal spiking was no longer observed during washout, and PADs could again be

evoked.

41

Figure 7. Evoked responses and tetanization-induced PADs in 2.5 mM K

+, 5 mM K

+, and

7.5 mM K+.

The stimulus strength for evoked responses was determined by an input-output curve and

the maximal field and population spike responses are shown in 2.5 mM K+ (A), 5 mM K

+

(B) and 7.5 mM K+ (C). There was no significant difference in the observed evoked

population spike and fEPSP amplitude among the different groups. Tetanization-induced

PADs were seen at both 2.5 mM K+ (D) and 5 mM K

+ (E). The PADs observed in 5 mM K

+

were significantly larger and longer-lasting than the PADs in 2.5 mM K+ (G,H). When

[K+]e was further elevated to 7.5 mM, no PAD could be induced by tetanization (F,G,H),

but there was tonic background activity resembling interictal spiking (F). The scale bar

(1 mV/50 ms) represents all evoked responses. The scale bar (1 mV/10 s) represents all

tetanization-induced PADs. * represents a significant difference from 2.5 mM K+ and **

represents a significant difference from 5 mM K+ (P<0.05).

42

6.4 The Effect of Blocking Ih on K+-induced attenuation of SLEs

After SLEs were spontaneously recurring in the low Mg2+

/5 mM K+ solution, adding 50

µM ZD7288 without an additional elevation in [K+]e did not significantly alter any of the

parameters. When ZD7288 was not included in the low Mg2+

/7.5 mM K+ solution, the

population spike and fEPSP amplitudes had significantly decreased to 46.3 ± 23.4% and 30.3 ±

15.2% of the control aCSF values (n=4) (P<0.05) (Figures 5A, 6A,B,8A,E). Furthermore, SLEs

were attenuated in the low Mg2+

/7.5 mM K+ (n=4) (Figures 5D,6C,8C,F). When 50 µM ZD7288

is included in the low Mg2+

/7.5 mM K+, recordings in the stratum radiatum revealed no

significant difference in the amplitude of the presynaptic volley (n=4). Recordings in the

pyramidal layer showed population spike and fEPSP amplitudes (105.2 ± 49.8% and 104.5 ±

19.2% respectively) similar to those recorded in control aCSF (n=4) (Figures 8B,E). Also,

distinct SLEs occurred quite frequently and were separated by short periods characterized by

interictal spiking activity (Figure 8D). The SLEs had amplitudes and durations (117.2 ± 43.1%

and 91.2 ± 32.5% respectively) similar to what was observed in low Mg2+

/5 mM K+ (Figure

8D,F).

43

Figure 8. The Effect of Blocking Ih on K+-induced Attenuation of SLEs.

The stimulus strength for evoked responses was determined by an input-output curve and the maximal field and population

spike responses are shown for low Mg2+

/7.5 mM K+

(A) and low Mg2+

/7.5 mM K+/ZD7288 (B). Including ZD7288 resulted in a

recovery of the population spike and fEPSP amplitude, back to levels seen in control aCSF (E). Low Mg2+

/7.5 mM K+

attenuated SLEs (C), but with 50 µM ZD7288, SLEs occurred frequently, and were separated by brief periods marked by

interictal activity (D). The SLEs had a similar duration and amplitude to values seen in low Mg2+

/5 mM K+ (F). The scale bar

(2 mV/50 ms) represents all evoked responses. The scale bar (2 mV/5 min) represents all spontaneous field recording

potentials. Asterisks denote a significant difference compared to low Mg2+

/7.5 mM K+ (P<0.05). In all parameters, the

addition of ZD7288 resulted in a significant increase in amplitude and duration (E,F).

44

7 Discussion

The most common form of epilepsy is complex partial epilepsy, which typically

originates in the temporal lobe (McNamara 1994). These seizures have been most commonly

studied in the hippocampal region, both in vitro and in vivo. It has been found that the seizure-

like events (SLEs) seen in the hippocampus are typically accompanied by an elevation in

extracellular potassium (K+

e). In healthy nervous tissue, [K+]e levels rarely reach 4.0 mM

(Somjen 1979), but during SLEs, [K+]e levels have been recorded to be up to 12 mM (Krnjevic et

al. 1982; Yaari et al. 1986). The main source of this [K+]e is assumed to be neuronal, and it has

been shown that neurons do release K+ with activity (Somjen 1979) but some have suggested

that there may be a large contribution of [K+]e from the astrocytic network (Bragin et al. 1997).

Many studies have since demonstrated that exposing tissue to elevated [K+]e (~8.5 mM) is

sufficient to induce SLEs in the hippocampus in vitro (Ogata 1975; Ogata 1976; Ogata et al.

1976; Traynelis and Dingledine 1988). There are various proposals as to why this may be. The

elevated [K+]e partially depolarizes the membrane and brings it closer to threshold (Alger et al.

1983). Since elevated [K+]e would decrease K

+ efflux, it has also been proposed that GABA-B

inhibitory postsynaptic potentials (IPSPs) would be decreased (Dutar and Nicoll 1988).

Furthermore, the burst afterhyperpolarization (AHP) which typically limits the amount of

repetitive firing possible is likewise decreased due to the diminished K+ efflux (Alger and Nicoll

1980). The decreased K+ efflux also will induce neuronal swelling, thereby reducing the

extracellular space, thus increasing the possibility of neuronal synchronization due to ephaptic

interactions (Dietzel et al. 1980).

The idea that accumulations in [K+]e may be what terminate seizures is something that

has been suggested, but at what point the K+ switches from being pro- to anti-convulsant has not

45

been examined before. In this study, the main objective was to examine whether elevated

[K+]e could attenuate in vitro epileptic activity in the rat hippocampus, and more specifically,

what is the range of [K+]e that would shift the action of K

+ from being pro- to anti-convulsant.

We then investigated whether this shift was strictly due to a depolarization block, as proposed in

past literature (Green 1964; Fertziger and Ranck 1970; Herreras and Somjen 1993; Bragin et al.

1997) , and also examined what other currents might be involved in this switch in the activity of

K+.

Here we used extracellular electrophysiological recordings in the rat hippocampus in

vitro during low Mg2+

/5 mM K+ -induced SLEs or tetanization-evoked PADs to show that

elevated [K+]e can indeed have anticonvulsant properties if elevated in epileptic tissue. We also

showed that this anticonvulsant behaviour is not simply a result of a depolarization block, but

other ionic currents, specifically the Ih, is involved in this switch.

Elevated [K+]e-induced Ih potentiation attenuates low Mg

2+/5 mM K

+ SLEs

Our study first confirmed that low Mg2+

/5 mM K+ does indeed produce robust seizure

activity, as is well characterized, including from various studies in our lab (Mody et al. 1987;

Armand et al. 2000; Derchansky et al. 2004; Isaev et al. 2005; Derchansky et al. 2006;

Derchansky et al. 2008). We then utilized this epileptic model to test the effects of elevated

[K+]e on neuronal excitability, as observed through the amplitude of the population spikes and

fEPSPs. Our results showed that there is a significant decrease in both population spike and

fEPSP amplitude when [K+]e is elevated to 7.5 mM and higher. However, contrary to the

depolarization block hypothesis, population spikes and fEPSPs, though attenuated, could still be

evoked, and recordings of the spontaneous potential revealed tonic background activity. This

suggests that the cells are still able to fire, so not all of the Na+ channels have been inactivated.

46

To assess whether this attenuation of the population spikes, fEPSPs, and SLEs is due to axonal

conduction failure, recordings were performed in the stratum radiatum to examine the effect of

the elevated [K+]e on the presynaptic volley, which would give an indication of the strength of

the afferent input. We found no significant difference in the presynaptic volley, which is in line

with observations made by Hablitz and Lundervold (1981), where they showed increasing [K+]e

from 3.25 to 12.25 mM in guinea pig hippocampal slices has less effect on the excitability of

presynaptic fibers (Hablitz and Lundervold 1981) than on postsynaptic neurons. This suggests

that the attenuation of SLEs is not a result of conduction failure down the axon.

To examine whether other ionic currents, specifically the Ih, are involved in the switch in

K+ activity, the Ih blocker ZD-7288 was applied to the bath solution. The idea that Ih may play a

role in terminating synchronous activity has been proposed earlier by Bal and McCormick, who

suggested that Ih conductance is responsible for the termination of synchronized oscillatory

activity in the thalamocortical network. As such, I examined whether Ih conductance in this

study may be involved in the termination of SLEs by elevated [K+]e.

The role of Ih in epilepsy has been widely debated, as different models have shown either

a pro- or an anti-convulsant effect (Di Pasquale et al. 1997; Poolos et al. 2002; Kitayama et al.

2003; Surges et al. 2003; Gill et al. 2006; Inaba et al. 2006). A recent review article examining

both sides and the potential for Ih to play a pro- or anti-convulsant role depending on the model

system suggests that the important factor is the balance between Ih pushing for excitation by

causing depolarizing, while simultaneously decreasing the input resistance of the tissue, thus

attenuating the effectiveness of incoming EPSPs (Dyhrfjeld-Johnsen et al. 2009). Ih is also

known to be highly influenced by [K+]e. Elevated [K

+]e has been shown to shift the reversal

potential and activation voltage of HCN channels to more depolarized levels, and even minor

47

fluctuations in [K+]e have been shown to drastically increase the amplitude of Ih going through

HCN channels (Spain et al. 1987; Funahashi et al. 2003; Shin and Carlen 2008). With this in

mind, we examined the effect of ZD7288 when [K+]e is further elevated in this epileptic model.

ZD7288 did not seem to have a pronounced effect on the evoked population spikes or

fEPSPs when added to low Mg2+

/5 mM K+. It also did not have a significant effect on the SLEs.

However, when ZD7288 was included in the perfusion of low Mg2+

/7.5 mM K+, a recovery of

population spike and fEPSP amplitude, along with SLEs, was observed. We propose, based on

our observations, that in this in vitro epileptic model, the elevated [K+]e increases the influence of

Ih, and it has an inhibitory effect by increasing the membrane conductance and decreasing the

membrane resistance, as others have suggested previously (Poolos et al. 2002). As such, when

[K+]e is elevated, there is a propagation failure of the signal up the dendrites due to the Ih-

induced decrease in input resistance, thus resulting in less EPSP summation and the observed

decrease in the evoked population spike and fEPSP amplitude. This is also in line with the

expression of HCN channels, which are far more populated in the dendritic area (Jung et al.

2007). This attenuation of EPSPs could also explain the presence of the low amplitude tonic

activity seen in the spontaneous potentials when [K+]e is elevated and seizures are attenuated.

With this proposal, it would suggest that when ZD7288 is applied, the Ih is blocked so input

resistance increases and the dendritic shunt is no longer active. As such, there is a recovery in

the population spike and fEPSP amplitude, as we observed. Also, with EPSPs no longer being

shunted and attenuated due to the increase in membrane resistance, the tissue becomes

hyperexcitable and SLEs occur again

48

Elevated [K+]e-induced Ih potentiation attenuates tetanization-induced PADs

To examine whether our finding regarding the anticonvulsant properties of elevated [K+]e

is generalizable to other in vitro models, we shifted from the low Mg2+

/5 mM K+ model and

examined the effects of elevated [K+]e on the PADs evoked by electrical stimulation, another

verified in vitro seizure model with a different mechanism of action. This model using electrical

stimulation is based off of the in vivo model of kindling, whereby repeated applications of

subconvulsive stimulus trains in the brain of a naïve animal will eventually result in ADs and

seizures (Stasheff et al. 1985). The advantage of this model over other chemically-induced

seizure models is that by only utilizing electrical stimulation, presumably the model should

mimic physiological communication between hyperexcitable areas of the brain, which is

necessary for the development of epilepsy. We found that by elevating [K+]e from control levels

of 2.5 mM up to 5 mM, there was a significant increase in the amplitude and duration of PADs.

This is in line with other literature which also uses 5 mM K+ in their control solution to promote

more robust PADs (Jahromi et al. 2002). The corresponding increase in amplitude and duration

of PADs resulting from the elevated [K+]e suggests that at this concentration, [K

+]e

is still

playing a pro-excitatory and pro-convulsant role, most likely by increasing the excitability of the

neurons by depolarizing the cells so they are closer to firing threshold, among other possible

mechanisms (Traynelis and Dingledine 1988). The further increase in [K+]e up to 7.5 mM

resulted in a blockade of PADs and instead, rhythmic interictal spiking activity was seen to

persist throughout the recording. This may also be due to a similar mechanism as seen in the low

Mg2+

/5 mM K+ model. The elevated [K

+]e may increase Ih and subsequently reduce tissue

resistance and attenuate EPSP summation. With the dendritic shunt, what would otherwise be

49

strong hyperexcitation traveling through the dendrites to the soma and producing SLEs or PADs,

may be reduced to a less excitable state in the form of the observed bursting activity.

Alternatively, recent modeling studies by Bazhenov et al., showed that increases in [K+]e

can result in the onset of an intrinsic burst mechanism (Bazhenov et al. 2004; Bazhenov et al.

2008). This intrinsic bursting mechanism is proposed to be due to elevations in [K+]e leading to

activation of other depolarizing currents, including Ih and INa,P, which pushes forward the next

burst. The limiting factor of these bursts is suggested to be the increased activation of the Ca2+

-

dependent K+ current, due to increased intracellular Ca

2+ resulting from the bursting. This

intrinsic bursting mechanism resulting from elevated [K+]e may be linked to a proposal by Avoli,

that interictal spiking activity may actually control seizure activity. He showed that interictal

activity originating in the hippocampal CA3 region can promote interictal spiking in the CA1

and prevent ictal activity (Avoli 2001). When this interictal activity was stopped by lesioning

the Schaffer collaterals, which are the CA3 input to the CA1, corresponding interictal spiking

stopped in the CA1, but seizure events began. When rhythmic electrical stimuli mimicking the

interictal activity was used, the ictal events seen in the CA1 were controlled again.

Here in this study, we show that the elevated [K+]e is accompanied by the tonic rhythmic

firing of a population of cells, and it is possible that the intrinsic bursting rhythm that

accompanies elevated [K+]e may be involved in the subsequent control of SLEs. As such, there

are two possibilities proposed here: that the intrinsic bursting rhythm seen to control seizures is a

consequence of the Ih-induced inhibition through decreasing membrane resistance and causing

EPSP propagation failure, or the elevated [K+]e potentiates the Ih to further drive the bursting

rhythm, and this bursting would then be the cause, rather than the consequence of seizure

attenuation.

50

8 Conclusions

1. There is a range of [K+]e at which the activity of K

+ switches from being pro- to anti-

convulsant (~6.5 mM to 12.5 mM).

2. The switch in K+ activity to being anticonvulsant is not simply due to depolarization

block, but involves a more complex system of interaction which involves the Ih.

3. 7.5 mM [K+]e appears to be an anticonvulsant [K

+]e for both the low Mg

2+/5 mM K

+ and

the tetanization models of epilepsy

4. The anticonvulsant activity of K+ is mediated, at least in part, by a potentiation of the Ih

5. The anticonvulsant properties of Ih is likely through a decrease in membrane resistance,

resulting in attenuation of incoming EPSPs

51

9 Ongoing/Future Work

This thesis has laid the groundwork for future studies examining the potential for

elevated [K+]e to exert an anticonvulsant influence through a potentiation of the Ih.

To further this project, the Ih blocker, ZD7288, needs to be tested in the tetanization

model to confirm that PADs will return even in elevated [K+]e as long as Ih is blocked. The use

of K+-sensitive electrodes would also lend strength to the story, as currently, bath perfusion of K

+

does not allow us to know what the actual [K+]e is in the deeper layers of the tissue.

Furthermore, the [K+]e is dynamic, especially during hyperexcited states such as during SLEs.

As such, K+-sensitive electrodes will allow for an understanding of how the K

+ dynamic is

actually shifting during the experimental procedure.

Intracellular recordings need to be done in the CA1 and CA3 pyramidal layers. The

resulting data will give an intracellular correlate to the extracellular field recordings we have

reported in this thesis. It would also lend more strength to this study if there were whole-cell

correlates in the interneurons as well, since HCN channels do become abundantly expressed in

the interneurons during development and interneurons play a crucial part in mediating

synchronous activity (Bender et al. 2001). Since HCN channels are differentially expressed

throughout the first few weeks of life, it would be interesting to see whether these anticonvulsant

properties of Ih hold throughout development.

Extracellular field recordings can also be performed in the CA3 pyramidal cell layer,

since it has been shown that the CA3 is the driver for SLEs whereas the CA1 is the receiver

(Dzhala and Staley 2003). This will allow for discernment with regards to whether the blockade

of SLEs at elevated K+ is due to a failure in SLEs propagating from CA3 to CA1, or whether

there is a failure to initiate SLEs in the CA3 region.

52

Other possible ion currents may be involved in this pro- to anti-convulsant switch in K+

activity, and one such candidate is the INa,P, as elevating [K+]e does enhance its conductance, and

attenuating INa,P has been shown to be anticonvulsant (Macdonald and Greenfield 1997) .

On a molecular level, it has been shown that some seizure models can produce cAMP

accumulation or upregulate adenosine receptors (Hattori et al. 1982; Hattori et al. 1993; Hattori

et al. 1993). This elevated cAMP would greatly potentiate Ih channels, so it would be beneficial

to run cAMP assays to see what levels of cAMP are seen during low Mg2+

/5 mM K+ or

tetanization seizure protocols and see whether the anticonvulsant properties are mediated in part

by a cAMP-dependent potentiation of Ih.

Ultimately, in vitro studies are limited so this model must be translated into an in vivo

model so that it has more clinical relevance. The efficacy of such a mechanism involving

elevated [K+]e and potentiation of Ih must then be tested, possibly through local injections of Ih

blockers or perhaps even deep brain stimulation protocols, whereby elevations in [K+]e would

result from the direct local high frequency stimulation of the brain.

53

10 References

1. Alger, B. E., et al. (1983). "On the possibility of simultaneously recording from two cells

with a single microelectrode in the hippocampal slice." Brain Res 270(1): 137-41.

2. Alger, B. E. and R. A. Nicoll (1980). "Epileptiform burst afterhyperolarization: calcium-

dependent potassium potential in hippocampal CA1 pyramidal cells." Science 210(4474): 1122-

4.

3. Amzica, F. and M. Steriade (2000). "Neuronal and glial membrane potentials during sleep

and paroxysmal oscillations in the neocortex." J Neurosci 20(17): 6648-65.

4. Armand, V., et al. (2000). "Effects of retigabine (D-23129) on different patterns of

epileptiform activity induced by low magnesium in rat entorhinal cortex hippocampal slices."

Epilepsia 41(1): 28-33.

5. Avoli, M. (2001). "Do interictal discharges promote or control seizures? Experimental

evidence from an in vitro model of epileptiform discharge." Epilepsia 42 Suppl 3: 2-4.

6. Awad, I. A., et al. (1991). "Intractable epilepsy and structural lesions of the brain: mapping,

resection strategies, and seizure outcome." Epilepsia 32(2): 179-86.

7. Awan, N. R., et al. (2009). "Deep brain stimulation: current and future perspectives."

Neurosurg Focus 27(1): E2.

8. Ayala, G. F., et al. (1970). "Excitability changes and inhibitory mechanisms in neocortical

neurons during seizures." J Neurophysiol 33(1): 73-85.

9. Ballanyi, K., et al. (1987). "Ion activities and potassium uptake mechanisms of glial cells in

guinea-pig olfactory cortex slices." J Physiol 382: 159-74.

10. Bazhenov, M., et al. (2008). "Cellular and network mechanisms of electrographic seizures."

Drug Discov Today Dis Models 5(1): 45-57.

11. Bazhenov, M., et al. (2004). "Potassium model for slow (2-3 Hz) in vivo neocortical

paroxysmal oscillations." J Neurophysiol 92(2): 1116-32.

12. Beaumont, V. and R. S. Zucker (2000). "Enhancement of synaptic transmission by cyclic

AMP modulation of presynaptic Ih channels." Nat Neurosci 3(2): 133-41.

13. Benardo, L. S. and D. A. Prince (1982). "Cholinergic excitation of mammalian hippocampal

pyramidal cells." Brain Res 249(2): 315-31.

14. Benardo, L. S. and D. A. Prince (1982). "Cholinergic pharmacology of mammalian

hippocampal pyramidal cells." Neuroscience 7(7): 1703-12.

15. Benardo, L. S. and D. A. Prince (1982). "Dopamine action on hippocampal pyramidal

cells." J Neurosci 2(4): 415-23.

16. Benardo, L. S. and D. A. Prince (1982). "Dopamine modulates a Ca2+-activated potassium

conductance in mammalian hippocampal pyramidal cells." Nature 297(5861): 76-9.

17. Benardo, L. S. and D. A. Prince (1982). "Ionic mechanisms of cholinergic excitation in

mammalian hippocampal pyramidal cells." Brain Res 249(2): 333-44.

18. Bender, R. A., et al. (2001). "Differential and age-dependent expression of

hyperpolarization-activated, cyclic nucleotide-gated cation channel isoforms 1-4 suggests

evolving roles in the developing rat hippocampus." Neuroscience 106(4): 689-98.

19. Benninger, C., et al. (1980). "Extracellular calcium and potassium changes in hippocampal

slices." Brain Res 187(1): 165-82.

20. Bialer, M., et al. (2001). "Progress report on new antiepileptic drugs: a summary of the Fifth

Eilat Conference (EILAT V)." Epilepsy Res 43(1): 11-58.

54

21. Bonnet, U., et al. (2000). "Alteration of intracellular pH and activity of CA3-pyramidal cells

in guinea pig hippocampal slices by inhibition of transmembrane acid extrusion." Brain Res

872(1-2): 116-24.

22. Boon, P. A., et al. (1991). "Intracranial, intraaxial, space-occupying lesions in patients with

intractable partial seizures: an anatomoclinical, neuropsychological, and surgical correlation."

Epilepsia 32(4): 467-76.

23. Borck, C. and J. G. Jefferys (1999). "Seizure-like events in disinhibited ventral slices of

adult rat hippocampus." J Neurophysiol 82(5): 2130-42.

24. Brager, D. H. and D. Johnston (2007). "Plasticity of intrinsic excitability during long-term

depression is mediated through mGluR-dependent changes in I(h) in hippocampal CA1

pyramidal neurons." J Neurosci 27(51): 13926-37.

25. Bragin, A., et al. (1997). "Termination of epileptic afterdischarge in the hippocampus." J

Neurosci 17(7): 2567-79.

26. Bragin, A., et al. (2002). "Rate of interictal events and spontaneous seizures in epileptic rats

after electrical stimulation of hippocampus and its afferents." Epilepsia 43 Suppl 5: 81-5.

27. Brewster, A. L., et al. (2005). "Formation of heteromeric hyperpolarization-activated cyclic

nucleotide-gated (HCN) channels in the hippocampus is regulated by developmental seizures."

Neurobiol Dis 19(1-2): 200-7.

28. Brown, D. A., et al. (1982). "Voltage-sensitive K-currents in sympathetic neurons and their

modulation by neurotransmitters." J Auton Nerv Syst 6(1): 23-35.

29. Brown, H. F., et al. (1979). "How does adrenaline accelerate the heart?" Nature 280(5719):

235-6.

30. Chamberlin, N. L. and R. Dingledine (1988). "GABAergic inhibition and the induction of

spontaneous epileptiform activity by low chloride and high potassium in the hippocampal slice."

Brain Res 445(1): 12-8.

31. Chen, K., et al. (2001). "Persistently modified h-channels after complex febrile seizures

convert the seizure-induced enhancement of inhibition to hyperexcitability." Nat Med 7(3): 331-

7.

32. Chevaleyre, V. and P. E. Castillo (2002). "Assessing the role of Ih channels in synaptic

transmission and mossy fiber LTP." Proc Natl Acad Sci U S A 99(14): 9538-43.

33. Cohen, I. and R. Miles (2000). "Contributions of intrinsic and synaptic activities to the

generation of neuronal discharges in in vitro hippocampus." J Physiol 524 Pt 2: 485-502.

34. Connors, B., et al. (1979). "LSD's effect on neuron populations in visual cortex gauged by

transient responses of extracellular potassium evoked by optical stimuli." Neurosci Lett 13(2):

147-50.

35. Connors, B. W., et al. (1982). "Electrophysiological properties of neocortical neurons in

vitro." J Neurophysiol 48(6): 1302-20.

36. Cooper, I. S., et al. (1973). "The effect of chronic cerebellar stimulation upon epilepsy in

man." Trans Am Neurol Assoc 98: 192-6.

37. Cortez, M. A., et al. (2001). "A model of atypical absence seizures: EEG, pharmacology,

and developmental characterization." Neurology 56(3): 341-9.

38. Davson, H. and M. B. Segal (1996). Physiology of the CSF and blood-brain barriers. Boca

Raton, CRC Press.

39. de Curtis, M., et al. (1999). "Cellular mechanisms underlying spontaneous interictal spikes

in an acute model of focal cortical epileptogenesis." Neuroscience 88(1): 107-17.

55

40. Dedeurwaerdere, S., et al. (2004). "Acute vagus nerve stimulation does not suppress spike

and wave discharges in genetic absence epilepsy rats from Strasbourg." Epilepsy Res 59(2-3):

191-8.

41. Derchansky, M., et al. (2008). "Transition to seizures in the isolated immature mouse

hippocampus: a switch from dominant phasic inhibition to dominant phasic excitation." J Physiol

586(2): 477-94.

42. Derchansky, M., et al. (2006). "Bidirectional multisite seizure propagation in the intact

isolated hippocampus: the multifocality of the seizure "focus"." Neurobiol Dis 23(2): 312-28.

43. Derchansky, M., et al. (2004). "Model of frequent, recurrent, and spontaneous seizures in

the intact mouse hippocampus." Hippocampus 14(8): 935-47.

44. Di Pasquale, E., et al. (1997). "Increased excitability and inward rectification in layer V

cortical pyramidal neurons in the epileptic mutant mouse Stargazer." J Neurophysiol 77(2): 621-

31.

45. Dichter, M. and W. A. Spencer (1969). "Penicillin-induced interictal discharges from the cat

hippocampus. II. Mechanisms underlying origin and restriction." J Neurophysiol 32(5): 663-87.

46. Dichter, M. A. and G. F. Ayala (1987). "Cellular mechanisms of epilepsy: a status report."

Science 237(4811): 157-64.

47. Dichter, M. A., et al. (1972). "Silent cells during interictal discharges and seizures in

hippocampal penicillin foci. Evidence for the role of extracellular K+ in the transition from the

interictal state to seizures." Brain Res 48: 173-83.

48. Dietzel, I., et al. (1980). "Transient changes in the size of the extracellular space in the

sensorimotor cortex of cats in relation to stimulus-induced changes in potassium concentration."

Exp Brain Res 40(4): 432-9.

49. DiFrancesco, D. (1986). "Characterization of single pacemaker channels in cardiac sino-

atrial node cells." Nature 324(6096): 470-3.

50. DiFrancesco, D., et al. (1986). "Properties of the hyperpolarizing-activated current (if) in

cells isolated from the rabbit sino-atrial node." J Physiol 377: 61-88.

51. DiFrancesco, D. and P. Tortora (1991). "Direct activation of cardiac pacemaker channels by

intracellular cyclic AMP." Nature 351(6322): 145-7.

52. Doan, T. N. and D. L. Kunze (1999). "Contribution of the hyperpolarization-activated

current to the resting membrane potential of rat nodose sensory neurons." J Physiol 514 ( Pt 1):

125-38.

53. Dutar, P. and R. A. Nicoll (1988). "A physiological role for GABAB receptors in the central

nervous system." Nature 332(6160): 156-8.

54. Dyhrfjeld-Johnsen, J., et al. (2008). "Upregulated H-Current in Hyperexcitable CA1

Dendrites after Febrile Seizures." Front Cell Neurosci 2: 2.

55. Dyhrfjeld-Johnsen, J., et al. (2009). "Double trouble? Potential for hyperexcitability

following both channelopathic up- and downregulation of Ih in epilepsy." Frontiers in

Neuroscience 3(1): 25-33.

56. Dzhala, V. I. and K. J. Staley (2003). "Transition from interictal to ictal activity in limbic

networks in vitro." J Neurosci 23(21): 7873-80.

57. Engel, J. (1989). Seizures and epilepsy. Philadelphia, F.A. Davis Co.

58. Fan, Y., et al. (2005). "Activity-dependent decrease of excitability in rat hippocampal

neurons through increases in I(h)." Nat Neurosci 8(11): 1542-51.

59. Feldberg, W. and S. L. Sherwood (1957). "Effects of calcium and potassium injected into

the cerebral ventricles of the cat." J Physiol 139(3): 408-16.

56

60. Fertziger, A. P. and J. B. Ranck, Jr. (1970). "Potassium accumulation in interstitial space

during epileptiform seizures." Exp Neurol 26(3): 571-85.

61. Fletcher, G. H. and V. A. Chiappinelli (1992). "An inward rectifier is present in presynaptic

nerve terminals in the chick ciliary ganglion." Brain Res 575(1): 103-12.

62. Funahashi, M., et al. (2003). "Role of the hyperpolarization-activated cation current (Ih) in

pacemaker activity in area postrema neurons of rat brain slices." J Physiol 552(Pt 1): 135-48.

63. Futamachi, K. J., et al. (1974). "Potassium activity in rabbit cortex." Brain Res 75(1): 5-25.

64. Gastaut, H. and R. Broughton (1972). Epileptic seizures; clinical and electrographic

features, diagnosis and treatment. Springfield, Ill., Thomas.

65. Gill, C. H., et al. (2006). "Inhibition of Ih reduces epileptiform activity in rodent

hippocampal slices." Synapse 59(5): 308-16.

66. Gotman, J. (1991). "Relationships between interictal spiking and seizures: human and

experimental evidence." Can J Neurol Sci 18(4 Suppl): 573-6.

67. Green, J. D. (1964). "The Hippocampus." Physiol Rev 44: 561-608.

68. Gutnick, M. J. and D. A. Prince (1981). "Dye coupling and possible electrotonic coupling in

the guinea pig neocortical slice." Science 211(4477): 67-70.

69. Hablitz, J. J. and A. Lundervold (1981). "Hippocampal excitability and changes in

extracellular potassium." Exp Neurol 71(2): 410-20.

70. Halliwell, J. V. and P. R. Adams (1982). "Voltage-clamp analysis of muscarinic excitation

in hippocampal neurons." Brain Res 250(1): 71-92.

71. Hattori, Y., et al. (1982). "Regional difference in cyclic AMP response to adenosine of rat

cerebral cortex with an iron-induced epileptic focus." Acta Med Okayama 36(4): 313-6.

72. Hattori, Y., et al. (1993). "Involvement of adenosine-sensitive cyclic AMP-generating

systems in cobalt-induced epileptic activity in the rat." J Neurochem 61(6): 2169-74.

73. Hattori, Y., et al. (1993). "Characterization of adenosine receptor-mediated generation of

cyclic AMP in slices of rat cerebral cortex with chronic epileptic activity." Neurochem Res

18(9): 1009-14.

74. Hauser, W. A., et al. (1993). "Incidence of epilepsy and unprovoked seizures in Rochester,

Minnesota: 1935-1984." Epilepsia 34(3): 453-68.

75. Heinemann, U. and H. D. Lux (1975). "Undershoots following stimulus-induced rises of

extracellular potassium concentration in cerebral cortex of cat." Brain Res 93(1): 63-76.

76. Herreras, O. and G. G. Somjen (1993). "Effects of prolonged elevation of potassium on

hippocampus of anesthetized rats." Brain Res 617(2): 194-204.

77. Herron, C. E., et al. (1986). "Frequency-dependent involvement of NMDA receptors in the

hippocampus: a novel synaptic mechanism." Nature 322(6076): 265-8.

78. Herron, C. E., et al. (1985). "A selective N-methyl-D-aspartate antagonist depresses

epileptiform activity in rat hippocampal slices." Neurosci Lett 61(3): 255-60.

79. Heuser, D., et al. (1975). "Brain carbonic acid acidosis after acetazolamide." Acta Physiol

Scand 93(3): 385-90.

80. Hodgkin, A. L., et al. (1952). "Measurement of current-voltage relations in the membrane of

the giant axon of Loligo." J Physiol 116: 424-448.

81. Howard, A. L., et al. (2007). "Opposing modifications in intrinsic currents and synaptic

inputs in post-traumatic mossy cells: evidence for single-cell homeostasis in a hyperexcitable

network." J Neurophysiol 97(3): 2394-409.

57

82. Inaba, Y., et al. (2006). "The H current blocker ZD7288 decreases epileptiform

hyperexcitability in the rat neocortex by depressing synaptic transmission." Neuropharmacology

51(3): 681-91.

83. Isaev, D., et al. (2005). "Anticonvulsant action of GABA in the high potassium-low

magnesium model of ictogenesis in the neonatal rat hippocampus in vivo and in vitro." J

Neurophysiol 94(4): 2987-92.

84. Jahromi, S. S., et al. (2002). "Anticonvulsant actions of gap junctional blockers in an in

vitro seizure model." J Neurophysiol 88(4): 1893-902.

85. Jami, L. (1972). "Patterns of cortical population discharges during metrazol-induced

seizures in cats." Electroencephalogr Clin Neurophysiol 32(6): 641-54.

86. Jensen, M. S. and Y. Yaari (1988). "The relationship between interictal and ictal paroxysms

in an in vitro model of focal hippocampal epilepsy." Ann Neurol 24(5): 591-8.

87. Jensen, M. S. and Y. Yaari (1997). "Role of intrinsic burst firing, potassium accumulation,

and electrical coupling in the elevated potassium model of hippocampal epilepsy." J

Neurophysiol 77(3): 1224-33.

88. Johnston, D. and T. H. Brown (1981). "Giant synaptic potential hypothesis for epileptiform

activity." Science 211(4479): 294-7.

89. Johnston, D., et al. (1980). "Voltage clamp discloses slow inward current in hippocampal

burst-firing neurones." Nature 286(5771): 391-3.

90. Jung, S., et al. (2007). "Progressive dendritic HCN channelopathy during epileptogenesis in

the rat pilocarpine model of epilepsy." J Neurosci 27(47): 13012-21.

91. Kaila, K. and B. R. Ransom (1998). pH and brain function. New York, Wiley-Liss.

92. Katona, I., et al. (1999). "Postsynaptic targets of somatostatin-immunoreactive interneurons

in the rat hippocampus." Neuroscience 88(1): 37-55.

93. Kitayama, M., et al. (2003). "Ih blockers have a potential of antiepileptic effects." Epilepsia

44(1): 20-4.

94. Knowles, W. D. and P. A. Schwartzkroin (1981). "Local circuit synaptic interactions in

hippocampal brain slices." J Neurosci 1(3): 318-22.

95. Konnerth, A. and U. Heinemann (1983). "Presynaptic involvement in frequency facilitation

in the hippocampal slice." Neurosci Lett 42(3): 255-60.

96. Korn, S. J., et al. (1987). "Epileptiform burst activity induced by potassium in the

hippocampus and its regulation by GABA-mediated inhibition." J Neurophysiol 57(1): 325-40.

97. Krnjevic, K., et al. (1982). "Stimulation-evoked changes in extracellular K+ and Ca2+ in

pyramidal layers of the rat's hippocampus." Can J Physiol Pharmacol 60(12): 1643-57.

98. Kwan, P. and M. J. Brodie (2000). "Early identification of refractory epilepsy." N Engl J

Med 342(5): 314-9.

99. Lebovitz, R. M., et al. (1971). "Recurrent excitation in the CA3 region of cat hippocampus."

Int J Neurosci 2(2): 99-107.

100. Lopantsev, V. and M. Avoli (1998). "Participation of GABAA-mediated inhibition in

ictallike discharges in the rat entorhinal cortex." J Neurophysiol 79(1): 352-60.

101. Ludwig, A., et al. (1998). "A family of hyperpolarization-activated mammalian cation

channels." Nature 393(6685): 587-91.

102. Lüders, H. and S. Noachtar (2000). Epileptic seizures : pathophysiology and clinical

semiology. New York, Churchill Livingstone.

103. Lux, H. D. (1974). "The kinetics of extracellular potassium: relation to epileptogenesis."

Epilepsia 15(3): 375-93.

58

104. Maccaferri, G. and C. J. McBain (1996). "The hyperpolarization-activated current (Ih) and

its contribution to pacemaker activity in rat CA1 hippocampal stratum oriens-alveus

interneurones." J Physiol 497 ( Pt 1): 119-30.

105. Macdonald, R. L. and L. J. Greenfield, Jr. (1997). "Mechanisms of action of new

antiepileptic drugs." Curr Opin Neurol 10(2): 121-8.

106. Macdonald, R. L. and K. M. Kelly (1995). "Antiepileptic drug mechanisms of action."

Epilepsia 36 Suppl 2: S2-12.

107. MacVicar, B. A. and F. E. Dudek (1980). "Dye-coupling betwen CA3 pyramidal cells in

slices of rat hippocampus." Brain Res 196(2): 494-7.

108. MacVicar, B. A. and F. E. Dudek (1980). "Local synaptic circuits in rat hippocampus:

interactions between pyramidal cells." Brain Res 184(1): 220-3.

109. Magee, J. C. (1999). "Dendritic Ih normalizes temporal summation in hippocampal CA1

neurons." Nat Neurosci 2(9): 848.

110. Magee, J. C. (2000). "Dendritic integration of excitatory synaptic input." Nat Rev

Neurosci 1(3): 181-90.

111. Magee, J. C. and M. Carruth (1999). "Dendritic voltage-gated ion channels regulate the

action potential firing mode of hippocampal CA1 pyramidal neurons." J Neurophysiol 82(4):

1895-901.

112. Mannikko, R., et al. (2002). "Voltage-sensing mechanism is conserved among ion

channels gated by opposite voltages." Nature 419(6909): 837-41.

113. Marcelin, B., et al. (2009). "h channel-dependent deficit of theta oscillation resonance and

phase shift in temporal lobe epilepsy." Neurobiol Dis 33(3): 436-47.

114. Matsumoto, H. and C. A. Marsan (1964). "Cortical Cellular Phenomena in Experimental

Epilepsy: Ictal Manifestations." Exp Neurol 9: 305-26.

115. Mayer, M. L. and G. L. Westbrook (1985). "The action of N-methyl-D-aspartic acid on

mouse spinal neurones in culture." J Physiol 361: 65-90.

116. McNamara, J. O. (1994). "Cellular and molecular basis of epilepsy." J Neurosci 14(6):

3413-25.

117. Meldrum, B. (2002). "Do preclinical seizure models preselect certain adverse effects of

antiepileptic drugs." Epilepsy Res 50(1-2): 33-40.

118. Mellor, J., et al. (2002). "Mediation of hippocampal mossy fiber long-term potentiation by

presynaptic Ih channels." Science 295(5552): 143-7.

119. Mirski, M. A. and R. S. Fisher (1994). "Electrical stimulation of the mammillary nuclei

increases seizure threshold to pentylenetetrazol in rats." Epilepsia 35(6): 1309-16.

120. Mody, I., et al. (1987). "Low extracellular magnesium induces epileptiform activity and

spreading depression in rat hippocampal slices." J Neurophysiol 57(3): 869-88.

121. Moosmang, S., et al. (1999). "Differential distribution of four hyperpolarization-activated

cation channels in mouse brain." Biol Chem 380(7-8): 975-80.

122. Moosmang, S., et al. (2001). "Cellular expression and functional characterization of four

hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues." Eur J Biochem

268(6): 1646-52.

123. Niedermeyer, E. (1972). The generalized epilepsies; a clinical electroencephalographic

study. With a foreword by A. Earl Walker. Springfield, Ill., C.C. Thomas.

124. Nolan, M. F., et al. (2007). "HCN1 channels control resting and active integrative

properties of stellate cells from layer II of the entorhinal cortex." J Neurosci 27(46): 12440-51.

59

125. Noma, A. and H. Irisawa (1976). "Membrane currents in the rabbit sinoatrial node cell as

studied by the double microelectrode method." Pflugers Arch 364(1): 45-52.

126. Ogata, N. (1975). "Ionic mechanisms of the depolarization shift in thin hippocampal

slices." Exp Neurol 46(1): 147-55.

127. Ogata, N. (1976). "Mechanisms of the stereotyped high-frequency burst in hippocampal

pyramidal neurons in vitro." Brain Res 103(2): 386-8.

128. Ogata, N., et al. (1976). "The correlation between extracellular potassium concentration

and hippocampal epileptic activity in vitro." Brain Res 110(2): 371-5.

129. Palva, J. M., et al. (2000). "Fast network oscillations in the newborn rat hippocampus in

vitro." J Neurosci 20(3): 1170-8.

130. Pape, H. C. (1996). "Queer current and pacemaker: the hyperpolarization-activated cation

current in neurons." Annu Rev Physiol 58: 299-327.

131. Pedley, T. A., et al. (1976). "Regulation of extracellular potassium concentration in

epileptogenesis." Fed Proc 35(6): 1254-9.

132. Pollo, C. and J. G. Villemure (2007). "Rationale, mechanisms of efficacy, anatomical

targets and future prospects of electrical deep brain stimulation for epilepsy." Acta Neurochir

Suppl 97(Pt 2): 311-20.

133. Poolos, N. P., et al. (2002). "Pharmacological upregulation of h-channels reduces the

excitability of pyramidal neuron dendrites." Nat Neurosci 5(8): 767-74.

134. Prince, D. A., et al. (1983). "Mechanisms underlying interictal-ictal transitions." Adv

Neurol 34: 177-87.

135. Proenza, C., et al. (2002). "Different roles for the cyclic nucleotide binding domain and

amino terminus in assembly and expression of hyperpolarization-activated, cyclic nucleotide-

gated channels." J Biol Chem 277(33): 29634-42.

136. Rafiq, A., et al. (1993). "Generation and propagation of epileptiform discharges in a

combined entorhinal cortex/hippocampal slice." J Neurophysiol 70(5): 1962-74.

137. Reiss, W. G. and K. S. Oles (1996). "Acetazolamide in the treatment of seizures." Ann

Pharmacother 30(5): 514-9.

138. Robinson, R. B. and S. A. Siegelbaum (2003). "Hyperpolarization-activated cation

currents: from molecules to physiological function." Annu Rev Physiol 65: 453-80.

139. Rutecki, P. A., et al. (1985). "Epileptiform activity induced by changes in extracellular

potassium in hippocampus." J Neurophysiol 54(5): 1363-74.

140. Saitow, F. and S. Konishi (2000). "Excitability increase induced by beta-adrenergic

receptor-mediated activation of hyperpolarization-activated cation channels in rat cerebellar

basket cells." J Neurophysiol 84(4): 2026-34.

141. Santoro, B., et al. (2004). "Regulation of HCN channel surface expression by a novel C-

terminal protein-protein interaction." J Neurosci 24(47): 10750-62.

142. Shah, M. M., et al. (2004). "Seizure-induced plasticity of h channels in entorhinal cortical

layer III pyramidal neurons." Neuron 44(3): 495-508.

143. Shin, D. S. and P. L. Carlen (2008). "Enhanced Ih depresses rat entopeduncular nucleus

neuronal activity from high-frequency stimulation or raised Ke+." J Neurophysiol 99(5): 2203-

19.

144. Shin, M., et al. (2008). "Mislocalization of h channel subunits underlies h channelopathy

in temporal lobe epilepsy." Neurobiol Dis 32(1): 26-36.

145. Shin, M. and D. M. Chetkovich (2007). "Activity-dependent regulation of h channel

distribution in hippocampal CA1 pyramidal neurons." J Biol Chem 282(45): 33168-80.

60

146. Singh, R., et al. (2002). "Chromosomal abnormalities and epilepsy: a review for clinicians

and gene hunters." Epilepsia 43(2): 127-40.

147. Smith, K. L., et al. (1995). "Diverse neuronal populations mediate local circuit excitation

in area CA3 of developing hippocampus." J Neurophysiol 74(2): 650-72.

148. Somjen, G. G. (1979). "Extracellular potassium in the mammalian central nervous

system." Annu Rev Physiol 41: 159-77.

149. Somjen, G. G. (2004). Ions in the Brain. New York, Oxford University Press, Inc.

150. Somjen, G. G., et al. (1985). "Sustained potential shifts and paroxysmal discharges in

hippocampal formation." J Neurophysiol 53(4): 1079-97.

151. Somjen, G. G., et al. (1986). "Interstitial ion concentrations and paroxysmal discharges in

hippocampal formation and spinal cord." Adv Neurol 44: 663-80.

152. Somjen, G. G. and J. L. Giacchino (1985). "Potassium and calcium concentrations in

interstitial fluid of hippocampal formation during paroxysmal responses." J Neurophysiol 53(4):

1098-108.

153. Somjen, G. G. and M. Muller (2000). "Potassium-induced enhancement of persistent

inward current in hippocampal neurons in isolation and in tissue slices." Brain Res 885(1): 102-

10.

154. Southan, A. P., et al. (2000). "Hyperpolarization-activated currents in presynaptic

terminals of mouse cerebellar basket cells." J Physiol 526 Pt 1: 91-7.

155. Spain, W. J., et al. (1987). "Anomalous rectification in neurons from cat sensorimotor

cortex in vitro." J Neurophysiol 57(5): 1555-76.

156. Stafstrom, C. E., et al. (1985). "Properties of persistent sodium conductance and calcium

conductance of layer V neurons from cat sensorimotor cortex in vitro." J Neurophysiol 53(1):

153-70.

157. Stafstrom, C. E., et al. (1982). "Negative slope conductance due to a persistent

subthreshold sodium current in cat neocortical neurons in vitro." Brain Res 236(1): 221-6.

158. Stafstrom, C. E., et al. (1984). "Properties of subthreshold response and action potential

recorded in layer V neurons from cat sensorimotor cortex in vitro." J Neurophysiol 52(2): 244-

63.

159. Stanfield, P. R., et al. (1985). "Substance P raises neuronal membrane excitability by

reducing inward rectification." Nature 315(6019): 498-501.

160. Stasheff, S. F., et al. (1985). "Induction of epileptiform activity in hippocampal slices by

trains of electrical stimuli." Brain Res 344(2): 296-302.

161. Stringer, J. L. and E. W. Lothman (1989). "Maximal dentate gyrus activation:

characteristics and alterations after repeated seizures." J Neurophysiol 62(1): 136-43.

162. Stringer, J. L., et al. (1989). "Induction of paroxysmal discharges in the dentate gyrus:

frequency dependence and relationship to afterdischarge production." J Neurophysiol 62(1): 126-

35.

163. Surges, R., et al. (2003). "Gabapentin increases the hyperpolarization-activated cation

current Ih in rat CA1 pyramidal cells." Epilepsia 44(2): 150-6.

164. Swartzwelder, H. S., et al. (1987). "Seizure-like events in brain slices: suppression by

interictal activity." Brain Res 410(2): 362-6.

165. Tombaugh, G. C. and G. G. Somjen (1996). "Effects of extracellular pH on voltage-gated

Na+, K+ and Ca2+ currents in isolated rat CA1 neurons." J Physiol 493 ( Pt 3): 719-32.

166. Traynelis, S. F. and R. Dingledine (1988). "Potassium-induced spontaneous electrographic

seizures in the rat hippocampal slice." J Neurophysiol 59(1): 259-76.

61

167. Ulens, C. and J. Tytgat (2001). "Functional heteromerization of HCN1 and HCN2

pacemaker channels." J Biol Chem 276(9): 6069-72.

168. van Hooft, J. A., et al. (2000). "Differential expression of group I metabotropic glutamate

receptors in functionally distinct hippocampal interneurons." J Neurosci 20(10): 3544-51.

169. Wahl-Schott, C. and M. Biel (2009). "HCN channels: structure, cellular regulation and

physiological function." Cell Mol Life Sci 66(3): 470-94.

170. Wainger, B. J., et al. (2001). "Molecular mechanism of cAMP modulation of HCN

pacemaker channels." Nature 411(6839): 805-10.

171. Walton, J. (1985). Brain's diseases of the nervous system. Oxford, Oxford University

Press.

172. Wierschke, S., et al. (2009). "Hyperpolarization-activated cation currents in human

epileptogenic neocortex." Epilepsia.

173. Wong, R. K., et al. (1986). "Cellular basis of neuronal synchrony in epilepsy." Adv Neurol

44: 583-92.

174. Woodbury, D. M. and J. W. Woodbury (1990). "Effects of vagal stimulation on

experimentally induced seizures in rats." Epilepsia 31 Suppl 2: S7-19.

175. Yaari, Y., et al. (1986). "Nonsynaptic epileptogenesis in the mammalian hippocampus in

vitro. II. Role of extracellular potassium." J Neurophysiol 56(2): 424-38.

176. Yamamoto, C., et al. (1980). "Potentiation of excitatory postsynaptic potentials during and

after repetitive stimulation in thin hippocampal sections." Exp Brain Res 38(4): 469-77.

177. Yu, F. H. and W. A. Catterall (2004). "The VGL-chanome: a protein superfamily

specialized for electrical signaling and ionic homeostasis." Sci STKE 2004(253): re15.

178. Yu, X., et al. (2007). "Calcium influx through If channels in rat ventricular myocytes." Am

J Physiol Cell Physiol 292(3): C1147-55.

179. Yu, X., et al. (2004). "Calcium influx through hyperpolarization-activated cation channels

(I(h) channels) contributes to activity-evoked neuronal secretion." Proc Natl Acad Sci U S A

101(4): 1051-6.

180. Zhang, K., et al. (2006). "Decreased IH in hippocampal area CA1 pyramidal neurons after

perinatal seizure-inducing hypoxia." Epilepsia 47(6): 1023-8.

181. Zhang, Y. F., et al. (1996). "Anticonvulsant drug effects on spontaneous thalamocortical

rhythms in vitro: ethosuximide, trimethadione, and dimethadione." Epilepsy Res 23(1): 15-36.

182. Zuckermann, E. C. and G. H. Glaser (1968). "Changes in hippocampal-evoked responses

induced by localized perfusion with high-potassium cerebrospinal fluid." Exp Neurol 22(1): 96-

117.

183. Zuckermann, E. C. and G. H. Glaser (1968). "Hippocampal epileptic activity induced by

localized ventricular perfusion with high-potassium cerebrospinal fluid." Exp Neurol 20(1): 87-

110.

184. Zuckermann, E. C. and G. H. Glaser (1968). "Pre- and post-convulsive changes in

hippocampal excitability induced by localized ventricular perfusion with high potassium CSF."

Trans Am Neurol Assoc 93: 79-83.